ABSTRACT
Accurately predicting the ridership of public-transit routes provides substantial benefits to both transit agencies, who can dispatch additional vehicles proactively before the vehicles that serve a route become crowded, and to passengers, who can avoid crowded vehicles based on publicly available predictions. The spread of the coronavirus disease has further elevated the importance of ridership prediction as crowded vehicles now present not only an inconvenience but also a public-health risk. At the same time, accurately predicting ridership has become more challenging due to evolving ridership patterns, which may make all data except for the most recent records stale. One promising approach for improving prediction accuracy is to fine-tune the hyper-parameters of machine-learning models for each transit route based on the characteristics of the particular route, such as the number of records. However, manually designing a machine-learning model for each route is a labor-intensive process, which may require experts to spend a significant amount of their valuable time. To help experts with designing machine-learning models, we propose a neural-architecture and feature search approach, which optimizes the architecture and features of a deep neural network for predicting the ridership of a public-transit route. Our approach is based on a randomized local hyper-parameter search, which minimizes both prediction error as well as the complexity of the model. We evaluate our approach on real-world ridership data provided by the public transit agency of Chattanooga, TN, and we demonstrate that training neural networks whose architectures and features are optimized for each route provides significantly better performance than training neural networks whose architectures and features are generic. © 2022 IEEE.
ABSTRACT
Testing is the main gap in the HIV continuum of care in France. Despite new guidelines and diversified HIV testing services, the total number of tests does not increase quickly enough to rapidly reduce the interval between infection and diagnosis. Since July 1<sup>st</sup>, 2019, the ALSO program offers a free HIV testing solution, without prescription, in any walk-in medical laboratory in Paris and the Alpes-Maritimes. It aims at improving access to HIV testing in addition to existing offers. The experimentation is backed by a multidimensional evaluation system. This article describes its implementation and the results of the first semester in terms of use and impact on the overall HIV testing activity in both regions. For the first six months of the experimentation, ALSO tests represent 8% of all the tests carried out in walk-in medical labs: in Paris 15,583 ALSO tests vs 175,938 prescription tests in 157 laboratories, in the Alpes-Maritimes 4,853 ALSO tests vs 54,082 prescription tests in 106 laboratories. The comparison between the second half of 2019 and the second half of 2020 shows a net increase in the volume of tests reimbursed by the national health insurance system. The information collected from public sexually transmitted infection (STI) clinics does not indicate any movement of their users to ALSO. The HIV positivity rates among ALSO tests (3.0 and 2.3/1,000 tests) are between those of prescription tests (2.0 and 1.6/1,000 tests) and those of public STI clinics (5.7 and 5.4/1,000 tests). Those encouraging interim results have led to the extension of the experiment until the end of 2020, in the Covid-19 crisis context.
ABSTRACT
The aims of this study were to evaluate the impact of the COVID-19 pandemic on emergency and elective hand surgery in four Italian regions that had either a high (Lombardy and Piemonte) or a low (Sicilia and Puglia) COVID-19 case load to discuss problems and to elaborate strategies to improve treatment pathways. A panel of hand surgeons from these different regions compared and discussed data from the centers they work in. The COVID-19 pandemic had an enormous impact on both elective and emergency surgery in Italy, not only in highly affected regions but also - and paradoxically even at a higher extent - in regions with a low COVID-19 case load. A durable and flexible redesign of hand surgery activities should be promoted, while changing and hopefully increasing human resources and enhancing administrative support. Telematics must also be implemented, especially for delivering rehabilitation therapy.