Accurate Estimation of Individual Transmission Rates Through Contact Analytics Using UWB Based Indoor Location Data

Effective control and prevention of infectious dis-eases often require detailed social network data categorized by individual movements and contact structures. Social network data is primarily accessible on a macro scale through sources such as RFID, Bluetooth, and mobile location data. This has limited our ability to accurately estimate transmission rates and identify significant patterns of disease spread within social gatherings. This study aims to address this gap by formulating a methodology that enables precise estimation of the disease transmission rates of each individual. Our methodology consists of the following steps (i) Installing a real-time location system and collection of indoor location data from two IRB-approved social events for the college students; (ii) Estimating infection probabilities as a function of distance between individuals calibrated with COVID-19 transmission data; (iii) Conducting a comprehensive analysis of contacts, including contact events, intensity, and duration; (iv) Utilizing the insights from the preceding steps to compute infection transmission rates for each participant. Our study revealed that the majority of contacts among participants are transient, typically lasting between 1 to 30 seconds within a proximity of 6 feet. However, infection transmission is associated with different types of contact durations and frequency, all hap-pening within a 6-foot area. Furthermore, a comparative analysis of the data from two indoor social events indicates a strong correlation between the duration of contacts and the nature of the social gathering (e.g., student orientation session versus informal meeting), alongside the density of participants (e.g., crowded versus sparsely populated environments). The proposed methodology is the first step toward accurately estimating in-dividual transmission rates, leveraging individual characteristics and the nature of the social event as key determinants. This will potentially enable epidemiologists and healthcare officials to tailor guidelines according to the unique dynamics of each social event and in an individual basis.