The article discusses new research from the University of Massachusetts Amherst that shows upgrading electrical and broadband infrastructure using a “dig once” approach is nearly 40% more cost-effective than replacing them separately. The study found that co-undergrounding – burying both electric and broadband internet lines together – saves costs, making it feasible for smaller towns in Massachusetts to make undergrounding upgrades.

Using computational modeling across various infrastructure upgrade scenarios, the researchers found that co-undergrounding is 39% more cost-effective than separately burying electrical and broadband wires. They also explored how aggressively towns should pivot to putting lines underground, considering factors such as the cost of converting lines from above ground to underground, the cost of outages, and the hours of outages that can be avoided if lines are underground.

A case study in Shrewsbury, Massachusetts, found that an aggressive co-undergrounding strategy over 40 years would cost $45.4 million but save $55.1 million from avoiding outages, considering factors like spoiled food, damaged home appliances, missed remote work hours, and increased use of backup power sources.

The researchers also took into account additional benefits such as increased property values from the aesthetic improvement of eliminating overhead lines, resulting in a net benefit of $11.3 million. They concluded that aggressively converting just electrical wires to underground was less expensive but had a significantly lower net benefit than co-undergrounding.

Future research directions include quantifying the impacts of co-undergrounding across various geographic locations and scenarios, investigating alternative underground routing options, and other potential outage mitigation strategies.

The study’s findings aim to help decision makers prioritize strategic planning for infrastructure upgrades, considering factors like soil composition, network type, and land use variables. The ultimate goal is to encourage utilities and towns to think strategically about upgrading electrical and broadband infrastructure using a “dig once” approach.

The world is facing a growing health crisis – drug-resistant infections. These infections are not only harder to treat but also require more expensive and toxic medications, leading to longer hospital stays and higher mortality rates. In 2021 alone, 450,000 people developed multidrug-resistant tuberculosis (TB), with treatment success rates dropping to just 57%, according to the World Health Organization.

Tulane University scientists have developed a groundbreaking artificial intelligence-based method that more accurately detects genetic markers of antibiotic resistance in deadly bacteria like TB and staph. This innovative approach has the potential to lead to faster and more effective treatments.

The researchers introduced a new Group Association Model (GAM) that uses machine learning to identify genetic mutations tied to drug resistance. Unlike traditional tools, which can mistakenly link unrelated mutations to resistance, GAM doesn’t rely on prior knowledge of resistance mechanisms, making it more flexible and able to find previously unknown genetic changes.

Current methods of detecting resistance take too long or miss rare mutations. Tulane’s model addresses both problems by analyzing whole genome sequences and comparing groups of bacterial strains with different resistance patterns to find genetic changes that reliably indicate resistance to specific drugs. This is like using the bacteria’s entire genetic fingerprint to uncover what makes it immune to certain antibiotics.

In the study, the researchers applied GAM to over 7,000 strains of Mtb and nearly 4,000 strains of S. aureus, identifying key mutations linked to resistance. They found that GAM not only matched or exceeded the accuracy of the WHO’s resistance database but also drastically reduced false positives, wrongly identified markers of resistance which can lead to inappropriate treatment.

The model’s ability to detect resistance without needing expert-defined rules means it could potentially be applied to other bacteria or even in agriculture, where antibiotic resistance is also a concern in crops. This tool can help us stay ahead of ever-evolving drug-resistant infections and provide a clearer picture of which mutations actually cause resistance, reducing misdiagnoses and unnecessary changes to treatment.

When combined with machine learning, the ability to predict resistance with limited or incomplete data improved. In validation studies using clinical samples from China, the machine-learning enhanced model outperformed WHO-based methods in predicting resistance to key front-line antibiotics. Catching resistance early can help doctors tailor the right treatment regimen before the infection spreads or worsens.

It’s vital that we stay ahead of ever-evolving drug-resistant infections. This AI-powered diagnosis tool has the potential to revolutionize the way we detect and treat these deadly bacteria, leading to better patient outcomes and improved global health.

Riding the crest of technological advancements, researchers at Osaka Metropolitan University have developed a groundbreaking machine learning-powered fluid simulation model that significantly reduces computation time without compromising accuracy. This innovative technique opens doors to potential applications in offshore power generation, ship design, and real-time ocean monitoring.

Predicting fluid behavior with precision is crucial for industries relying on wave and tidal energy, as well as for designing maritime structures and vessels. Traditional particle methods are commonly used but require extensive computational resources. The new AI-powered surrogate model uses graph neural networks to simplify and accelerate fluid simulations, making waves in fluid dynamics research.

However, researchers acknowledge that AI is not without its limitations. “AI can deliver exceptional results for specific problems but often struggles when applied to different conditions,” said Takefumi Higaki, an assistant professor at Osaka Metropolitan University’s Graduate School of Engineering and lead author of the study.

The team aimed to create a tool that is consistently fast and accurate. They first compared different training conditions to determine what factors were essential for high-precision fluid calculations. Then, they systematically evaluated how well their model adapted to different simulation speeds and various types of fluid movements.

The results demonstrated strong generalization capabilities across different fluid behaviors. “Our model maintains the same level of accuracy as traditional particle-based simulations, throughout various fluid scenarios, while reducing computation time from approximately 45 minutes to just three minutes,” Higaki said.

This research marks a significant step forward in high-performance fluid simulation, offering a scalable and generalizable solution that balances accuracy with efficiency. Such improvements extend beyond the lab, enabling faster and more precise fluid simulations that can accelerate the design process for ships and offshore energy systems.

The study was published in Applied Ocean Research, highlighting the potential of AI to revolutionize ocean simulations and unlock new possibilities for industries reliant on wave and tidal energy.