Modeling pneumococcal resistance to penicillin in southern Sweden using artificial neural networks.

In recent decades, penicillin-resistant pneumococci (PRP) have emerged and spread rapidly between and within countries over the world. In this study we developed an iterative artificial neural network (ANN) model to describe and predict the spread of PRP in space and time as a function of antibiotic consumption and a number of different confounders. Retrospective data from 1997 to 2000 on an international epidemic PRP clone (serotype 9V) and antibiotic consumption data from Southern Sweden were used to train the ANN models and data from 2001 to 2003 for evaluation of the model predictions. Five different ANN models were trained, each with independent topology optimization for alternative sets of input variables to find the most descriptive model. The model containing all variables was the only one performing better than the reference linear models, as assessed by the correlation between predictions and observations. The inability to identify a smaller subset of most predictive parameters may reflect either diffuse causal mechanisms or just the absence of critical experimental indicators from the dataset. The iterative ANN model identified is useful to predict future data. The sensitivity analysis of the model suggests that past incidence has a small effect on the number of PRP cases.

Implementing a public web based GIS service for feedback of surveillance data on communicable diseases in Sweden.

BACKGROUND: Surveillance data allow for analysis, providing public health officials and policy-makers with a basis for long-term priorities and timely information on possible outbreaks for rapid response (data for action). In this article we describe the considerations and technology behind a newly introduced public web tool in Sweden for easy retrieval of county and national surveillance data on communicable diseases. METHODS: The web service was designed to automatically present updated surveillance statistics of some 50 statutory notifiable diseases notified to the Swedish Institute for Infectious Disease Control (SMI). The surveillance data is based on clinical notifications from the physician having treated the patient and laboratory notifications, merged into cases using a unique personal identification number issued to all Swedish residents. The web service use notification data from 1997 onwards, stored in a relational database at the SMI. RESULTS: The web service presents surveillance data to the user in various ways; tabulated data containing yearly and monthly disease data per county, age and sex distribution, interactive maps illustrating the total number of cases and the incidence per county and time period, graphs showing the total number of cases per week and graphs illustrating trends in the disease data. The system design encompasses the database (storing the data), the web server (holding the web service) and an in-the-middle computer (to ensure good security standards). CONCLUSIONS: The web service has provided the health community, the media, and the public with easy access to both timely and detailed surveillance data presented in various forms. Since it was introduced in May 2003, the system has been accessed more than 1,000,000 times, by more than 10,000 different viewers (over 12.600 unique IP-numbers).