ABSTRACT
Agent-based modeling and simulation techniques are widely and successfully used for analyzing complex and emergent phenomena in many research and application areas. Among the many different reasons which sustain the flexibility and success of such techniques, it is important to mention the availability of a great variety of software tools, easing (1) the development of models, (2) the execution of simulations, and (3) the analysis of results. Currently, with the rapid global spread of the COVID-19 pandemic, one of the most important research area is dedicated to define algorithms and systems to support epidemic forecasting simulations, scalable on large populations. In particular, in this paper, we propose an agent-based epidemic model and a distributed architecture that can be used for the simulation of populations represented by millions of agents. Moreover, the paper presents the results of the simulations on the data of the population of Lombardy. © 2021 CEUR-WS. All rights reserved.
ABSTRACT
Background: The COVID-19 global pandemic required a great organizational effort to reduce the n. of patient contacts in Hospital. Based on a recent Italian survey 80% of cancer centers adopted alternative modalities to get in touch with patients. However there are no data on the remote management in follow-up. Material and methods: In this study we collected data regarding oncological patients in follow-up whose control visit was scheduled between 11/03 and 15/05 at AOUC. Categorical data were analysed by chi square or Fisher exact test;statistical analysis of continuous variables was performed by Mann Whitney U test. Results: 222 patients were enrolled: 99 man, 123 women (median age 68 yrs). The 43.6% were affected by gastrointestinal tumors, 21.1% genitourinary, 17.5% gynecological/ breast, 8.5% thoracic and 7.2% melanoma. 35% were stage I, 30% stage II, 23% stage III and 12% stage IV. The median time from diagnosis was 3 years. 192 out of 222 (86%) did not underwent the planned medical examination;28% of them postponed (9% managed by call, 1 by email and 89 both). 51% of them were contacted through the caregiver. The % of patients that underwent medical examination was 4.4, 11.7 e 24.25 in March, April and May respectively, suggesting a different compliance with respect to remote management, correlated to a perception of the risk of infection. The median postponement time of was 28 days (5-51). For almost all patients (95.5%) there was no evidence of disease, for 2.7% was registered a relapse and for 1.8% additional examination was ongoing: a correlation was observed between in-person visit and the recurrence (p=0.012) and between elderly patients and caregiver mediated contact (p=0.002). Conclusions: This experience shows that the remote management of cancer patients in follow-up is feasible. Many aspects need to be clarified: lates outcomes, patient satisfaction, type of patients who can benefit. Almost all patients were managed both by telephonic interview and by e-mail, resulting in longer consultation time. Likely, a preventive and exhaustive patient information and a better technological equipment would improve the quality and the duration of the tele-consultation. The incremental percentage of in-person visits is another aspect to be investigated. Lastly, greater attention and training should be addressed to caregivers. The correlation emerged between in person in-visit and recurrence could be explained by a proper physician screening.