Epidemic exposure risk assessment in digital contact tracing: A fuzzy logic approach.

Background: Bluetooth low energy (BLE)-based contact-tracing applications were widely used during the COVID-19 pandemic. However, the use of only the received signal strength feature for proximity calculations may not be adaptable to different virus variants or scalable for other potential epidemic diseases. Objective: This study presents a novel framework in regard to evaluating and classifying personal exposure risk that considers both contact features, which include distance and length of contact, and environment features, which include crowd size and the number of recently infected cases in the environment. The framework utilizes a fuzzy expert system that is adaptable to different virus variants. Methods: The proposed method was tested on two viruses with different close contact features, which used four membership functions and 256 fuzzy rule sets. Results: The proposed framework classified personal exposure risks into four classes, which include low, medium, high, and too high risk. The empirical results showed that the fuzzy logic-based approach reduced the number of false positive cases and demonstrated better accuracy and precision than the current BLE-only approaches. Conclusions: The proposed framework provides a more practical and adaptable method in regard to assessing exposure risks in real-world scenarios. It has the potential to be scalable and adaptable to different virus variants and other potential epidemic diseases by considering both contact and environment features. These findings may be useful in order to develop more effective digital contact-tracing applications and policies.