Auflistung nach Autor:in "Hilfiker, Lorenz"

Gerade angezeigt 1 - 2 von 2

Active querying approach to epidemic source detection on contact networks
(Nature, 2023) Sterchi, Martin; Hilfiker, Lorenz; Grütter, Rolf; Bernstein, Abraham
The problem of identifying the source of an epidemic (also called patient zero) given a network of contacts and a set of infected individuals has attracted interest from a broad range of research communities. The successful and timely identification of the source can prevent a lot of harm as the number of possible infection routes can be narrowed down and potentially infected individuals can be isolated. Previous research on this topic often assumes that it is possible to observe the state of a substantial fraction of individuals in the network before attempting to identify the source. We, on the contrary, assume that observing the state of individuals in the network is costly or difficult and, hence, only the state of one or few individuals is initially observed. Moreover, we presume that not only the source is unknown, but also the duration for which the epidemic has evolved. From this more general problem setting a need to query the state of other (so far unobserved) individuals arises. In analogy with active learning, this leads us to formulate the active querying problem. In the active querying problem, we alternate between a source inference step and a querying step. For the source inference step, we rely on existing work but take a Bayesian perspective by putting a prior on the duration of the epidemic. In the querying step, we aim to query the states of individuals that provide the most information about the source of the epidemic, and to this end, we propose strategies inspired by the active learning literature. Our results are strongly in favor of a querying strategy that selects individuals for whom the disagreement between individual predictions, made by all possible sources separately, and a consensus prediction is maximal. Our approach is flexible and, in particular, can be applied to static as well as temporal networks. To demonstrate our approach’s practical importance, we experiment with three empirical (temporal) contact networks: a network of pig movements, a network of sexual contacts, and a network of face-to-face contacts between residents of a village in Malawi. The results show that active querying strategies can lead to substantially improved source inference results as compared to baseline heuristics. In fact, querying only a small fraction of nodes in a network is often enough to achieve a source inference performance comparable to a situation where the infection states of all nodes are known.
01A - Beitrag in wissenschaftlicher Zeitschrift
Covid-19 superspreading. lessons from simulations on an empirical contact network
(2021) Hilfiker, Lorenz; Sterchi, Martin
Infectious individuals who cause an extraordinarily large number of secondary infections are colloquially referred to as superspreaders. Their pivotal role for the transmission of Covid-19 has been exemplified by now infamous cases such as the Washington choir practice, where one infectious individual caused 52 secondary infections [1]. In order to formally analyse superspreading, we denote by Z the individual reproduction number. In a fully susceptible population, the mean mZ is known as the basic reproduction number R0. Based on branching arguments and assuming a well-mixed population, the distribution of Z is typically modelled by a negative binomial distribution whose variance mZ(1+mZ=kZ) is characterised by the dispersion parameter kZ [2]. Empirical evidence suggests that Covid-19 exhibits a particularly wide distribution of Z, with the right tail representing superspreading events. In situations without interventions, the dispersion parameter kZ was estimated in the range 0.04 - 0.2 [3, 4]. Some studies even found evidence for a fat tailed Z-distribution, possibly a power law with the exponent close to 1 [5, 6]. The underlying mechanisms for the emergence of this level of heterogeneity are difficult to establish. A priori, network effects could play a role, as suggested in [5]. A more frequent line of reasoning focuses on physiological or biological factors: wet pronunciation, loud speech, frequent coughing or higher viral loads could result in some infected individuals being inherently more prone to spread the disease than others during an encounter with a susceptible individual [7]. Combining both lines of thought, the study in [8] shows that individual variation in infectiousness indeed leads to higher variance of Z on some standard static network models. However, no previous study has investigated heterogeneities of the Z-distribution on empirical contact networks. Therefore, we provide preliminary simulation results based on one realistic temporal social contact network and gather further evidence that the key to finding Z-distributions in alignment with empirical data is to allow for individual variation in infectiousness.
06 - Präsentation