Latest news with #GradientBoosting
_2025_06_17_14_16_55.jpg&w=3840&q=100)
GMA Network
17-06-2025
- Health
- GMA Network
UP researchers use AI models to predict antimicrobial resistance
Researchers from the University of the Philippines have tapped artificial intelligence to predict antimicrobial resistance, particularly in agricultural environments. UP researchers used Escherichia coli (E. coli) for testing antimicrobial resistance since it can easily develop resistance to antibiotics. E. coli is a common bacterium that inhabits the intestines of animals and humans and is often used to identify fecal contamination. UP researchers tested AI prediction models to determine the antimicrobial resistance of E. coli using genetic data and laboratory test results from the National Center for Biotechnology Information (NCBI) database. These AI models are Random Forest (RF), Support Vector Machine (SVM), and two ensemble methods—Adaptive Boosting (AB) and Extreme Gradient Boosting (XGB). 'We selected the models based on their strengths in handling biological and imbalanced data,' said Dr. Pierangeli Vital of UP Diliman College of Science's Natural Sciences Research Institute. 'These models were chosen to compare performance across different learning strategies and to identify which is most suitable for predicting antibiotic resistance,' she added. The study showed that the AI models most accurately predicted resistance to streptomycin and tetracycline, both types of antibiotics. However, ciprofloxacin, another type of antibiotic, was the most challenging to predict due to the limited number of resistant samples in the data (only 4%), which led to difficulty in identifying resistance and poor sensitivity. The study noted that AB and XGB consistently delivered good results, even when tested on imbalanced antimicrobial resistance data. 'We think that this strategy has great potential for real-time monitoring of antimicrobial resistance, particularly in agriculture,' Vital said. 'As DNA sequencing becomes faster and cheaper, prediction models such as ours can pick up resistant bacteria early—before they lead to outbreaks. This can facilitate better decision-making in food safety, agriculture, and public health programs,' she added. The researchers recommend including more diverse sample types and data sources to better understand and predict how bacteria develop resistance. The study titled 'Prediction models for the antimicrobial resistance of Escherichia coli in an agricultural setting around Metro Manila, Philippines' was published in the Malaysian Journal of Microbiology. —VBL, GMA Integrated News
Yahoo
04-03-2025
- Science
- Yahoo
Scientists make major breakthrough while testing new solar technology outdoors — here's what this means for the future of energy
Solar panels play a major role in helping us achieve our sustainability goals by harnessing the sun's abundant energy and converting it into clean electricity. Perovskite-based photovoltaic cells have shown great promise for increased efficiency and lower costs over traditional silicon versions, but they tend to suffer from long-term stability issues. There's also been a lack of comprehensive outdoor testing to prove their viability in the real world — until now. Researchers from Belgium's Interuniversity Microelectronics Centre and the University of Cyprus have announced the completion of two-year outdoor stability tests involving perovskite solar mini-modules that they developed, as PV Magazine reported, and the results are promising. "The most durable mini-module maintained 78% of its initial efficiency after one year," the researchers shared in the report. There was a nominal drop of 7-8% efficiency during the initial burn-in period, but it soon leveled out. Stress factors that tend to affect these solar panels include moisture, UV light exposure, and temperature, but according to the study, much of the previous testing had been done in controlled indoor environments. There's also a phenomenon involving diurnal and nighttime changes, wherein there's been degradation during daylight hours but also overnight recovery of solar modules. The research team included these factors into their equation, along with a focus on the evolution of key electrical parameters and the performance of entire modules instead of just individual cells. The report says that these elements were overlooked by other studies. "The results have also been disseminated and discussed at conferences and with our partners in different projects, which comprise both academia and industry. There is high interest since the data sets are some of the most extensive ones currently available for outdoor testing of perovskite modules," Tom Aernouts, research and development manager at IMEC, Hasselt University, and EnergyVille, said. The information they gathered was also used to develop a data-driven predictive model to gauge power output by using the eXtreme Gradient Boosting regression technique, as the report noted. It used the normalized root mean square error to gauge how well the model matches observed data. If you were to install home solar panels, which of these factors would be your primary motivation? Energy independence Lower power bills Helping the planet No chance I ever go solar Click your choice to see results and speak your mind. "Our predictive model, focusing on essential environmental parameters, accurately forecasted the power output of mini-modules with a 6.76% nRMSE, indicating its potential to predict the lifetime of perovskite-based devices," the researchers wrote. Developing more affordable, efficient, and resilient photovoltaic modules could help supercharge the solar industry as well as consumer adoption. Plus, a detailed understanding of the long-term operational capabilities of perovskite models will only bolster further study, helping the energy industry transition away from dirty fuels and invest in greener technologies. Join our free newsletter for weekly updates on the latest innovations improving our lives and shaping our future, and don't miss this cool list of easy ways to help yourself while helping the planet.
