Abstract
The massive volume of hemodynamic data routinely available within the Cardiovascular Intensive Care Unit (CVICU) can adversely affect the quality, relevance and timing of hemodynamic management decisions on patients after cardiac surgery. Yet, at the same time, the lack of appropriate treatment-outcome data and access to prior CV case histories deprives the clinician of any opportunity to improve personal decision-making skill and assess the effectiveness of various treatment methods. This paper presents a formalized decision-support model for CVICU that incorporates expert and quantitative knowledge, as well as prior outcome and case experience to augment the clinician's decision-making capability. This includes the proposed use of optimal hemodynamic patterns derived from outcome analysis as therapy goals, expert rules and trend analysis to interpret incoming data, standardized protocols based on predefined hemodynamic patterns from clinical cases, and access to the database for similar case comparison. Most importantly, the model suggests an integrated approach where the clinical database is not only a documentation source for the patient, but can also serve as an outcome research database where clinical experience can be formalized and combined with expert knowledge to influence future therapy decisions. At present, a prototype is being developed at the CVICU of the University of Alberta Hospitals on a Unix platform using ART-IM, C and Ingres. Once implemented, the prototype will be evaluated on a small group of CV patients for its effectiveness and acceptability to clinicians.
Full text
PDF
Selected References
These references are in PubMed. This may not be the complete list of references from this article.
- Langlotz C. P., Fagan L. M., Tu S. W., Sikic B. I., Shortliffe E. H. A therapy planning architecture that combines decision theory and artificial intelligence techniques. Comput Biomed Res. 1987 Jun;20(3):279–303. doi: 10.1016/0010-4809(87)90059-0. [DOI] [PubMed] [Google Scholar]
- Miller R. A. INTERNIST-1/CADUCEUS: problems facing expert consultant programs. Methods Inf Med. 1984 Jan;23(1):9–14. [PubMed] [Google Scholar]