Anomaly detection system for toxicology protocol studies

✍ Scribed by Mark W. Naylor; Robert B. Weiner; Aaron F. Levy; Roger M. Folk

• Publisher: Elsevier Science
• Year: 1980
• Tongue: English
• Weight: 829 KB
• Volume: 13
• Category: Article
• ISSN: 0010-4809
• DOI: 10.1016/0010-4809(80)90044-0

✦ Synopsis

Large-scale longitudinal toxicologic animal studies necessarily generate a considerable volume of numerical data points, though only a relatively small number of these data may be variations of interest to the investigator. This computer-based method detects these anomalies of interest by the use of predetermined criteria which create an individual element "window" to which each data point is compared. The anomalies thus detected are presented in various formats which assist the investigator in drawing scientific conclusions. The tedious manual task of screening the original data is eliminated, the accuracy is improved and professional time is conserved. The methodology of setting the windows is described, and typical sample outputs are presented.