Rule-based anomaly pattern detection for detecting disease outbreaks

This paper presents an algorithm for performing early detection of disease outbreaks by searching a database of emergency department cases for anomalous patterns. Traditional techniques for anomaly detection are unsat-isfactory for this problem because they identify individ-ual data points that are rare due to particular combina-tions of features. When applied to our scenario, these traditional algorithms discover isolated outliers of par-ticularly strange events, such as someone accidentally shooting their ear, that are not indicative of a new out-break. Instead, we would like to detect anomalous pat-terns. These patterns are groups with specific character-istics whose recent pattern of illness is anomalous rel-ative to historical patterns. We propose using a rule-based anomaly detection algorithm that characterizes each anomalous pattern with a rule. The significance of each rule is carefully evaluated using Fisher’s Exact Test and a randomization test. Our algorithm is com-pared against a standard detection algorithm by mea-suring the number of false positives and the timeliness of detection. Simulated data, produced by a simulator that creates the effects of an epidemic on a city, is used for evaluation. The results indicate that our algorithm has significantly better detection times for common sig-nificance thresholds while having a slightly higher false positive rate.