Browse Prior Art Database

PATTERN OPTIMIZATION METHOD FOR PREVENTIVE MAINTENANCE IN MEDICAL IMAGING SYSTEMS

IP.com Disclosure Number: IPCOM000109661D
Publication Date: 2005-Mar-24
Document File: 6 page(s) / 35K

Publishing Venue

The IP.com Prior Art Database

Abstract

In one embodiment, a pattern optimization method for preventive maintenance in medical imaging systems includes generating patterns containing all parameters from medical equipments and using the generated patterns to create a model to predict maintenance of a set of medical equipments. At least one pattern is dropped from the generated set for creating the model. If the absence of the dropped pattern gives the same result as that with the pattern included, then the pattern becomes a candidate for elimination from the optimal set of patterns. If absence of this pattern does not produce the same result as that with the pattern included, then the pattern is considered as an essential element and marked as undeletable.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 36% of the total text.

PATTERN OPTIMIZATION METHOD FOR PREVENTIVE MAINTENANCE IN MEDICAL IMAGING SYSTEMS

FIELD OF THE INVENTION

[0001]   This invention relates generally to optimization methods and, more particularly, to a pattern optimization method for preventive maintenance in medical imaging systems.

BACKGROUND OF THE INVENTION 

[0002]     Companies around the globe are becoming increasingly dependent on the income derived from servicing the equipments that they sell or third party equipments.  The servicing of the equipments involves periodic maintenance and reactive repairs.  The servicing is usually done on a periodic basis and may result in over servicing or under servicing of equipments.  Over servicing of equipments may result in lost revenue. Under servicing may lead to customer productivity loss.  The service offerings can become more attractive to companies if the service resources are used effectively to provide optimal and proactive offerings to equipments.

[0003]   The proactive and optimal offerings require the use of parametric machine data from equipments to develop a prediction engine and feeding it continuously and automatically to analyze the trend of equipment-critical data. Therefore, parametric data has potential to attain very high importance in the servicing of medical equipments.  The extraction of useful knowledge using knowledge discovery can provide indicators of the health of the medical equipments and open the door for machine data driven equipment servicing.  This will result in savings from unnecessary repair and part replacement.  The proactive and timely repair will reduce the equipment downtime resulting in customer productivity, patient comfort, lower medical cost and contribution towards bottom line of medical industry.  This untapped knowledge can be regarded as the currency of the service industry.

[0004]   The knowledge discovery finds previously unknown, implicit, potentially useful and non-trivial patterns and regularities in the machine data. Usually it is very difficult to identify the patterns, which are interesting to the domain in which they are going to be applied for making a prediction. There has been work in other areas than medical industries to determine the interesting patterns or a complete set of patterns.

[0005]   However, the number of parameters involved in any particular class of equipments is very large and requires very expensive computing. Sometimes, these parameters are also conflicting with each other.  This may also need extensive amount of storage of redundant and useless machine data to be stored online. 

[0006]   Thus, there exists a need to determine an optimal set of patterns, which can provide prediction with acceptable certainty in a cost effective and timely manner.  This optimal set has to be dynamically derived to serve different types of equipments or different fleet of equipments within same type of equipments.

SUMMARY OF THE INVENTION

[0007]    In one embodiment, a pattern optimizati...