Signal processing for monitoring and diagnosis: Medical applications

✍ Scribed by Suzanne Lesecq; Sylvie Charbonnier

• Publisher: John Wiley and Sons
• Year: 2010
• Tongue: English
• Weight: 45 KB
• Volume: 24
• Category: Article
• ISSN: 0890-6327
• DOI: 10.1002/acs.1150

✦ Synopsis

System monitoring and diagnosis is a research area that has received a growing interest from both the signal processing community and the control community in the recent years. Indeed, the development of new sensors makes it possible to record variables on a system that were not acquired previously. These new data provide useful information on the inner evolution of the system. Moreover, complex numerical treatment can be easily performed due to the computational capabilities and storage capacities of today processors and computers. Fed by this information, supervision systems aim at detecting abnormal situations so as to increase the system safety. Several applications areas, such as sewer networks [1], mining industries and mineral flotation plants [2] or aeronautics [3], have been benefited from the recent advances in the diagnosis field.

In medicine, the system monitored is a patient, which makes the notion of safety all the more important. Therefore, reliability is a strong requirement for a medical monitoring system. It must detect promptly any degradation in the patient's state and avoid false alarms. Note that missed detections obviously decrease the physician's reliance, whereas false alarms decrease