Wearable health technology is drawing significant attention for good reasons. The pervasive nature of such systems providing ubiquitous access to the continuous personalized data will transform the way people interact with each other and their environment. The resulting information extracted from these systems will enable emerging applications in healthcare, wellness, emergency response, fitness monitoring, elderly care support, long-term preventive chronic care, assistive care, smart environments, sports, gaming, and entertainment which create many new research opportunities and transform researches from various disciplines into data science which is the methodological terminology for data collection, data management, data analysis, and data visualization. Despite the ground-breaking potentials, there are a number of interesting challenges in order to design and develop wearable medical embedded systems. Due to limited available resources in wearable processing architectures, power-efficiency is demanded to allow unobtrusive and long-term operation of the hardware. Also, the data-intensive nature of continuous health monitoring requires efficient signal processing and data analytic algorithms for real-time, scalable, reliable, accurate, and secure extraction of relevant information from an overwhelmingly large amount of data. Therefore, extensive research in their design, development, and assessment is necessary. Embedded Processing Platform Design The majority of my work concentrates on designing wearable embedded processing platforms in order to shift the conventional paradigms from hospital-centric healthcare with episodic and reactive focus on diseases to patient-centric and home-based healthcare as an alternative segment which demands outstanding specialized design in terms of hardware design, software development, signal processing and uncertainty reduction, data analysis, predictive modeling and information extraction. The objective is to reduce the costs and improve the effectiveness of healthcare by proactive early monitoring, diagnosis, and treatment of diseases (i.e. preventive) as shown in Figure 1.
Our group works at the forefront of deploying large-scale sensor networks to the built environment for monitoring and control of civil infrastructure systems including bridges, roads, rail networks, and pipelines; this research portfolio falls within the broader class of cyber-physical systems (CPS). To maximize the benefit of the massive data sets we collect from operational infrastructure systems, we undertake research in the area of relational and NoSQL database systems, cloud-based analytics, and data visualization technologies. In addition, our algorithmic work is focused on the use of statistical signal processing, pattern classification, machine learning, and model inversion/updating techniques to automate the interrogation sensor data collected. The ultimate aim of our work is to harness the full potential of data science to provide system users with real-time, actionable information obtained from the raw sensor data collected.
Prof. Vershynin’s main area of expertise is high dimensional probability and its applications. He is interested in random geometric structures that appear in various data science problems. The following is a sample of his recent projects: 1. High dimensional inference from nonlinear data Sometimes we are given certain observations of an unknown vector that encodes useful but hidden information, and we want to compute that vector. Examples includes compressed sensing, linear and non-linear regression, as well as binary (yes-no) observations. We are developing methods that can estimate the hidden vector without even knowing the nature of the non-linearity of observations. Areas of application include survey methodologies, signal processing, and various high-dimensional classification problems. 2. Structure mining in networks Complex data sets such as networks often have latent structures, for example clusters or communities. We are interested in developing efficient methods to discover such latent structures. Prof. Vershynin’s methods come from various areas of mathematics and data science, including random matrix theory, geometric functional analysis, convex and discrete geometry, geometric combinatorics, high dimensional statistics, information theory, learning theory, signal processing, theoretical computer science and numerical analysis.
Professor Christopher Miller is an observational cosmologist who works in the fields of astronomical data mining and computational astrostatistics. He co-founded the INternational Computational Astrostatistics (INCA) group, a collaboration of researchers from the University of Michigan, Carnegie Mellon University, University of Washington, Georgia Tech, the NOAO, and others. Recently, he led the NOAO Science Data Management group, where he was responsible for using and delivering science quality astronomical data from instruments like the MOSAIC optical and NEWFIRM IR images on NOAO’s 4m-class telescopes. He was hired at the University of Michigan under a U-M Presidential initiative for advancing data mining research. His research and teaching emphasizes open source collaborative code development and the use of cloud computing to analyze large volumes of astronomical data. Professor Miller’s group develops and applies advanced computational and statistical techniques to address the following research areas:
- Cosmological parameter inference using the abundance and spatial distribution of clusters of galaxies.
- Evolution of the physical properties of the brightest galaxy cluster members.
- Morphological classification of galaxies.
- Dynamical techniques to trace the gravitational potential of galaxy clusters and probe the theory of gravity over large scales.
- Advanced data reduction pipelines for multi-object spectroscopic data.
- Concurrent real-world and simulated data analysis