Stream Query Processing for Healthcare Bio-sensor Applications
ICDE '04 Proceedings of the 20th International Conference on Data Engineering
Design, implementation, and evaluation of the linear road bnchmark on the stream processing core
Proceedings of the 2006 ACM SIGMOD international conference on Management of data
SPC: a distributed, scalable platform for data mining
Proceedings of the 4th international workshop on Data mining standards, services and platforms
Real-time analysis of physiological data to support medical applications
IEEE Transactions on Information Technology in Biomedicine
Two Heads Better Than One: Metric+Active Learning and its Applications for IT Service Classification
ICDM '09 Proceedings of the 2009 Ninth IEEE International Conference on Data Mining
Body sensor data processing using stream computing
Proceedings of the international conference on Multimedia information retrieval
A System for Mining Temporal Physiological Data Streams for Advanced Prognostic Decision Support
ICDM '10 Proceedings of the 2010 IEEE International Conference on Data Mining
Intensive Care Window: Real-Time Monitoring and Analysis in the Intensive Care Environment
IEEE Transactions on Information Technology in Biomedicine
Hi-index | 0.00 |
There is a tremendous amount of data available to physicians at the point of care in intensive care environments; however, physicians do not have the tools to extract relevant clinical information in a timely manner. They mostly rely on manual inspection of the data to make diagnosis and prognosis. New software technologies make it possible to automatically generate meaningful information in real-time from the physiological data streams of patients. These real-time monitoring software technologies can support multiple concurrent patients and have been developed mainly to be applied in a reactive way, for the detection of patient complications. This paper proposes ways to extend these real-time monitoring technologies to help intensive care become more proactive. We present a system design and algorithms for a prototype system that produces in real-time short-term predictions of patient physiological data from live and historical patient data. One technique is based solely on the patient's own live data streams. The other technique is based on comparing the patient's physiological data streams with data streams of similar patients that have been monitored in the past. An extensive experimental study of this system is proposed to evaluate its predictive ability.