Using video-analysis technology to estimate social mixing and simulate influenza transmission at a mass gathering.

Mass gatherings create settings conducive to infectious disease transmission. Empirical data to model infectious disease transmission at mass gatherings are limited. Video-analysis technology could be used to generate data on social mixing patterns needed for simulating influenza transmission at mass gatherings. We analyzed short video recordings of persons attending the GameFest event at a university in Troy, New York, in April 2013 to demonstrate the feasibility of this approach. Attendees were identified and tracked during three randomly selected time periods using an object-tracking algorithm. Tracks were analyzed to calculate the number and duration of unique pairwise contacts. A contact occurred each time two attendees were within 2 m of each other. We built and tested an agent-based stochastic influenza simulation model assuming two scenarios of mixing patterns in a geospatially accurate representation of the event venue -one calibrated to the mean cumulative contact duration estimated from GameFest video recordings and the other using a uniform mixing pattern. We compared one-hour attack rates (i.e., becoming infected) generated from these two scenarios following the introduction of a single infectious seed. Across the video recordings, 278 attendees were identified and tracked, resulting in 1,247 unique pairwise contacts with a cumulative mean contact duration of 74.76 s (SD: 80.71). The one-hour simulated mean attack rates were 2.17 % (95 % CI:1.45 - 2.82) and 0.21 % (95 % CI: 0.14 - 0.28) in the calibrated and uniform mixing model scenarios, respectively. We simulated influenza transmission at the GameFest event using social mixing data objectively captured through video-analysis technology. Microlevel geospatially accurate simulations can be used to assess the layout of event venues on social mixing and disease transmission. Future work can expand on this demonstration project to larger spatial and temporal scenes in more diverse settings.

Factorization-Based Texture Segmentation.

This paper introduces a factorization-based approach that efficiently segments textured images. We use local spectral histograms as features, and construct an M × N feature matrix using M-dimensional feature vectors in an N-pixel image. Based on the observation that each feature can be approximated by a linear combination of several representative features, we factor the feature matrix into two matrices--one consisting of the representative features and the other containing the weights of representative features at each pixel used for linear combination. The factorization method is based on singular value decomposition and nonnegative matrix factorization. The method uses local spectral histograms to discriminate region appearances in a computationally efficient way and at the same time accurately localizes region boundaries. The experiments conducted on public segmentation data sets show the promise of this simple yet powerful approach.

Estimating contact rates at a mass gathering by using video analysis: a proof-of-concept project.

BACKGROUND: Current approaches for estimating social mixing patterns and infectious disease transmission at mass gatherings have been limited by various constraints, including low participation rates for volunteer-based research projects and challenges in quantifying spatially and temporally accurate person-to-person interactions. We developed a proof-of-concept project to assess the use of automated video analysis for estimating contact rates of attendees of the GameFest 2013 event at Rensselaer Polytechnic Institute (RPI) in Troy, New York. METHODS: Video tracking and analysis algorithms were used to estimate the number and duration of contacts for 5 attendees during a 3-minute clip from the RPI video. Attendees were considered to have a contact event if the distance between them and another person was ≤1 meter. Contact duration was estimated in seconds. We also simulated 50 attendees assuming random mixing using a geo-spatially accurate representation of the same GameFest location. RESULTS: The 5 attendees had an overall median of 2 contact events during the 3-minute video clip (range: 0-6). Contact events varied from less than 5 seconds to the full duration of the 3-minute clip. The random mixing simulation was visualized and presented as a contrasting example. CONCLUSION: We were able to estimate the number and duration of contacts for 5 GameFest attendees from a 3-minute video clip that can be compared to a random mixing simulation model at the same location. The next phase will involve scaling the system for simultaneous analysis of mixing patterns from hours-long videos and comparing our results with other approaches for collecting contact data from mass gathering attendees.