‘This Program Provides the Opportunity to Do Cutting-Edge Research’

Can Ye: “This Program Provides the Opportunity to Do Cutting-Edge Research”

Can Ye Can Ye started his dual degree doctoral program in Electrical and Computer Engineering (ECE) in the fall of 2008 at the Universidade do Porto and Carnegie Mellon University. He selected the Carnegie Mellon Portugal Program because of the excellent opportunities of research collaboration, and because of the exciting experience that he is gaining in Europe, which he thinks may be of great help to his future career.

Passionate by biomedical signal processing, pattern recognition and machine learning, Can Ye is currently working on the Vital Responder Project. The Vital Responder Project aims at realizing real-time health monitoring of first responders (e.g., fire fighters) wearing Vital Jacket, a smart textile capable of collecting various vital signals via embedded miniature biosensors, including electrocardiogram (ECG) signal, respiration and blood pressure. He is the primary research investigator in the signal processing group of the project in the Carnegie Mellon Portugal program, funded by the Fundação para a Ciência e a Tecnologia. This project involves several Portuguese universities (Universidade de Aveiro, Universidade do Porto), research institutes (IEETA, IT), companies (Biodevices, Petratex, McLaren Electronics), and Carnegie Mellon University.

Can Ye doctoral thesis aims to “develop advanced signal processing and machine learning systems to enable the intelligent interpretation – processing, modeling and mining – of vital signals such as ECG signals, blood pressure and respiration, collected via wearable miniature sensors.” He thinks that such system can provide a comprehensive estimation of the user’s real-time health status, so as to detect health abnormalities and alarming trends autonomously, and consequently, warn the user and inform health care givers for appropriate action. He explained that this system “combined with superior bio-sensing and bio-communication technologies can realize ubiquitous real-time health monitoring for various populations, such as elderly and chronic disease patients.”

Since the beginning of his Ph.D., Can Ye accomplished “three IEEE conference publications, one journal, and there are two conference submissions underway,” with his advisors Miguel Coimbra, from the Faculdade de Ciências da Universidade do Porto (FCUP) and Instituto de Telecomunicações (IT), and Vijayakumar Bhagavatula, from Carnegie Mellon. The first publication titled “Arrhythmia Detection and Classification using Morphological and Dynamic Features of ECG Signals” (Proc. IEEE EMBC, Buenos Aires, Argentina, Sep 2010), “resulted in the best performance (99.59%) so far in the literature for detecting and classifying arrhythmias via ECG signals on the baseline MIT-BIH Arrhythmias Database,” said Can Ye. The MIT-BIH Arrhythmias Database is regarded as the benchmark database in the topic of cardiac arrhythmias detection and classification.

Recently Can Ye, J. Pallauf (CMU), B.V.K. Vijaya Kumar (CMU), and M. Coimbra (FCUP), presented the paper “Customizing Training Dataset for Improved Heartbeat Recognition Performance in Long-Term ECG Signal Analysis”, at the IEEE Engineering in Medicine and Biology Society (EMBC) on August 30 – September 3, 2011, in Boston, MA (USA) . In this paper the authors presented “an investigation of the potential benefits of customizing the analysis of long-term ECG signals, collected from individuals using wearable sensors, by incorporating small amount of data from these individuals in the training set of our classifiers.” One of the conclusions is that “given that the scenario of long-term, real-time and ubiquitous health monitoring in unsupervised environments is concerned, our study hints that the long-term ECG signal analysis requires frequent adaptation of the pre-determined global settings to tackle both the individual and context differences.” For this reason, the authors said that “in future work, it will be desirable to develop a self-adaptable classifier to cope with these variations.”

Can Ye feels that “this novel program provides not only the excellent opportunity to do cutting-edge research in two excellent research institutions in the area, but also the chance to experience and savor different cultures in the US and in Europe.” He believes “this unique cross-continent experience will prove beneficial to my future professional career.”

September 2011