DocumentCode
3365166
Title
An intelligent assistant for sleep stage scoring
Author
Bentrup, John A.
Author_Institution
Illinois Univ., Urbana, IL, USA
fYear
1992
fDate
14-17 Jun 1992
Firstpage
325
Lastpage
334
Abstract
An intelligent assistant for sleep stage scoring is currently being developed. Due to the complexities and variations in the sleep signals, it is quite common that the system will initially be unable to correctly classify portions of the sleep signals. However, the system works in conjunction with somnologists, or human sleep stage scoring experts, and actively seeks their advice when it is unable to reliably decide on its own. The system incorporates the human guidance into its own knowledge base and continues with its scoring. The emphasis is on finding ways to produce reliable classifications in a domain that is at times very difficult to categorize. This reliability is critical if the physicians, in reaching their diagnoses and recommendations, are to rely on the machine´s output. The system has successfully scored 156 test cases with an overall 92% epoch-by-epoch agreement with the human expert
Keywords
medical expert systems; medical signal processing; pattern recognition; advice; diagnoses; human guidance; intelligent assistant; knowledge base; reliable classifications; sleep signals; sleep stage scoring; somnologists; Biomedical electrodes; Computer displays; Electroencephalography; Electromyography; Electrooculography; Humans; Medical diagnostic imaging; Pattern analysis; Sleep; System testing;
fLanguage
English
Publisher
ieee
Conference_Titel
Computer-Based Medical Systems, 1992. Proceedings., Fifth Annual IEEE Symposium on
Conference_Location
Durham, NC
Print_ISBN
0-8186-2742-5
Type
conf
DOI
10.1109/CBMS.1992.244932
Filename
244932
Link To Document