Pulsar Algorithm: Syndromic Surveillance Data from Athens Hospitals, 2002-2003

A research paper evaluates the Pulsar algorithm for detecting outbreaks in syndromic time series data. The analysis uses daily syndromic counts from emergency departments of four major hospitals in the Athens area during August 2002 to August 2003. The work was authored by Urania Dafni of the University of West Attica.

Use Cases

• Evaluating peak detection algorithms based on syndromic time series data mentioned in the description
• Simulating outbreak detection performance based on sensitivity and specificity criteria described in the paper
• Analyzing trends and noise in public health surveillance data based on the described methodology
• Comparing statistical methods for timeliness in alerting based on the described simulation schema

Strengths

• The description details a specific simulation schema used for method evaluation
• Analysis is based on data from four major hospitals in the Athens area
• The method is evaluated on performance criteria of sensitivity, specificity, and timeliness

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download
• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment

Provenance

Source

Urania Dafni, University of West Attica

Collection Method

Analysis of daily syndromic counts from hospital emergency departments, with simulations performed to evaluate the Pulsar method.

Time Range

August 2002 to August 2003

Geography

Athens, Greece