The emergence of antibiotic-resistant pathogenic bacteria is a major concern for human health, rendering some antibiotics ineffective in treating diseases. The authors of this study tested the hypothesis that exposing rumen bacteria to tetracycline will gradually lead to the development of tetracycline-resistant bacteria, some of which will develop multidrug resistance.

In the field of ecology, little is known about how plant communities originate. Through the process of characterizing dunes, mounds of sand formed by the wind, and their plant communities we can get to know the physiognomy and floristic composition of the territory. Based on the hypothesis that dune flora can emerge from seed islands: holes in the sand 6 cm deep containing a mixture of seeds, broken branches of shrubbery, and rabbit feces, during spring, the authors determined the composition of 20 seed islands in the sand dunes of Concon, Chile and measured how many seeds germinated in each one.

Seeking to investigate the effects of ambient pollutants on human respiratory health, here the authors used machine learning to examine asthma in Lost Angeles County, an area with substantial pollution. By using machine learning models and classification techniques, the authors identified that nitrogen dioxide and ozone levels were significantly correlated with asthma hospitalizations. Based on an identified seasonal surge in asthma hospitalizations, the authors suggest future directions to improve machine learning modeling to investigate these relationships.

While serving as an immediate address for psychological safety and stability, psychological first aid (PFA) currently lacks the incorporation of triage. Without triage, patients cannot be prioritized in correspondence to condition severity that is often called for within emergency conditions. To disentangle the relevance of a potential triage system to PFA, the authors of this paper have developed a method to quantify resilience - a prominent predictor of the capability to recover from a disaster. With this resilience index, they have quantified resilience of differing age, race, and sex demographics to better inform the practice of PFA and potential demographic prioritization via a triage system.

Using facial recognition as a use-case scenario, we attempt to identify sources of bias in a model developed using transfer learning. To achieve this task, we developed a model based on a pre-trained facial recognition model, and scrutinized the accuracy of the model’s image classification against factors such as age, gender, and race to observe whether or not the model performed better on some demographic groups than others. By identifying the bias and finding potential sources of bias, his work contributes a unique technical perspective from the view of a small scale developer to emerging discussions of accountability and transparency in AI.

The traditional alert system in California consists of Wireless Emergency Alerts (WEAs), which lack location specificity, and sign-up-based technology which is limited by the number of sign ups. Those who do not have phones or have a silence option on their devices are most at risk from the current alert system. Here the authors developed cloud-enabled crisis connection for disaster alerts (CRISIS-CONNECT) to mitigate problems associated with the current alert system.

One of the greatest challenges we face today is the sustainable production, storage, and distribution of electrical power. One emerging technology with great promise in this area is that of metal-air fuel cells—a long-term and reusable electricity storage system made from a reactive metal anode and a saline solution. In this study the authors tested several different types of metal to determine which was the most suitable for this application. They found that a fuel cell with a magnesium anode was superior to fuel cells made from aluminum or zinc, producing a voltage and current sufficient for real-world applications such as charging a mobile phone.

The authors designed a system that runs off of body heat to track body temperature that could help prevent injuries that result from elevated body temperature.

Regenerative medicine has become a mainstay in recent times, and employing stem cells to treat several degenerative, inflammatory conditions has resulted in very promising outcomes. These forms of cell-based therapies are novel approaches to existing treatment modalities. In this study, the authors compared the concentrations of the cytokines PDGF, IL-8, and VEGF between conditioned and spent media of mesenchymal stem cells (MSCs) to evaluate their potential therapeutic properties for wound healing in inflammatory conditions. They hypothesized that conditioned media contains higher concentrations of wound healing cytokines compared to spent media. The authors found that while IL-8 and VEGF were present in highest concentrations in conditioned media, PDGF was present in maximal amounts in spent media.

In this paper, we measured the privacy budgets and utilities of different differentially private mechanisms combined with different machine learning models that forecast traffic congestion at future timestamps. We expected the ANNs combined with the Staircase mechanism to perform the best with every value in the privacy budget range, especially with the medium high values of the privacy budget. In this study, we used the Autoregressive Integrated Moving Average (ARIMA) and neural network models to forecast and then added differentially private Laplacian, Gaussian, and Staircase noise to our datasets. We tested two real traffic congestion datasets, experimented with the different models, and examined their utility for different privacy budgets. We found that a favorable combination for this application was neural networks with the Staircase mechanism. Our findings identify the optimal models when dealing with tricky time series forecasting and can be used in non-traffic applications like disease tracking and population growth.