Browse Articles

This study investigates skin tone bias in artificial intelligence models used for dermatological disease classification and evaluates a CycleGAN-based data augmentation approach to improve diagnostic performance on darker skin types. We generated synthetic dark-skinned images to enhance dataset diversity and compared model performance before and after augmentation. The results demonstrate that augmentation with synthetic dermatological images can help reduce disparities in diagnostic performance across skin tones, highlighting a practical strategy for improving fairness in dermatology AI systems.

This manuscript explores the use of Spanish language translators in an outpatient clinic in New Jersey. The authors surveyed patients before and after in person, video, and telephone appointments to determine which modality was most acceptable to the patients. The authors found that the three modalities did not differ in patient satisfaction, but that patients were grateful for translations services and that patient trust may be expanded by the use of these services and by focuses of translator soft skills.

This paper discusses how the conditions at the center of stars affects the nuclear reactions that happen inside these stars. We focused on the effect of temperature and density and found that these two properties interacted to create a greater effect when combined than when separate.

This study investigates the ability of Gymnema sylvestre tea to acutely suppress sweet taste perception across various food groups. Our findings demonstrate how this natural botanical can reduce sugar cravings, offering a potential complementary strategy for managing dietary intake in type 2 diabetes.

The authors looked at how different blends of synthetic and non-synthetic fabrics protected against UV radiation.

This study examined whether findings from laboratory studies of baseball cognition also apply during real games. Using 2024 MLB data from 30 players, researchers analyzed batting performance under different levels of feedback and anxiety.

This study investigated perceptions of the emerging workforce toward the use of artificial intelligence in hiring, specifically for assessing subjective "culture fit." Through a mixed-methods survey of 150 high school and early-college students in Nepal, we found a significant disconnect between organizational adoption of AI and the profound skepticism of young job candidates, who express deep concerns about fairness, transparency, and the potential for AI to perpetuate systemic discrimination.

This study shows the efficacy of leveraging transfer learning, specifically from residual networks, to detect CVDs and possible signs of CVDs. The findings indicate that leveraging transfer learning from residual networks alongside medical professionals is a highly promising approach for CVD detection and diagnosis, warranting further investigation.

Excess sucrose consumption has been associated with several health problems, including inflammation and potential negative effects on immune function. However, the exact relationship between sucrose intake and immunity remains unclear, especially during bacterial infections. This study examined how sucrose intake affected the survival of fruit flies following oral infection with the bacterial pathogen Serratia marcescens.

Heavy metal contamination from transportation can damage soil and vegetation by increasing oxidative stress, disrupting enzymes, and impairing photosynthesis. This study investigated whether chondroitin sulfate (CS) could reduce the harmful effects of contaminated roadside soil on plant growth by binding with metal ions.