Science Advances is the American Association for the Advancement of Science’s (AAAS) open access multidisciplinary journal, publishing impactful research papers and reviews in any area of science, in both disciplinary-specific and broad, interdisciplinary areas. The mission of Science Advances is to provide fair, fast, and expert peer review to authors and a vetted selection of research, freely available to readers. Led by a team of distinguished scientists and allowing flexible article formats, Science Advances supports the AAAS mission by extending the capacity of Science magazine to identify and promote significant advances in science and engineering across a wide range of areas. The journal’s use of evolving digital publishing technologies plays a critical role in building and sustaining AAAS’s mission as a global participant and advocate for the communication and use of science to benefit humankind.
Roxana Daneshjou, Kailas Vodrahalli, Weixin Liang, Melissa J. Jenkins, Veronica Rotemberg, Justin Ko, Susan M. Swetter, Elizabeth E. Bailey, Olivier Gevaert, Pritam Mukherjee, Michelle Phung, Kiana Yekrang, Bradley Fong, Rachna Sahasrabudhe, James Zou, Albert S. Chiou
An estimated 3 billion people lack access to dermatological care globally. Artificial intelligence (AI) may aid in triaging skin diseases and identifying malignancies. However, most AI models have not been assessed on images of diverse skin tones or uncommon diseases. Thus, we created the Diverse Dermatology Images (DDI) dataset—the first publicly available, expertly curated, and pathologically confirmed image dataset with diverse skin tones. We show that state-of-the-art dermatology AI models exhibit substantial limitations on the DDI dataset, particularly on dark skin tones and uncommon diseases. We find that dermatologists, who often label AI datasets, also perform worse on images of dark skin tones and uncommon diseases. Fine-tuning AI models on the DDI images closes the performance gap between light and dark skin tones. These findings identify important weaknesses and biases in dermatology AI that should be addressed for reliable application to diverse patients and diseases.
Soumya Mukherjee, Nivedita Sikdar, Daniel O’Nolan, Douglas Franz, Victoria Gascón, Amrit Kumar, Naveen Kumar, Hayley S. Scott, David G. Madden, Paul E. Kruger, Brian Space, Michael J. Zaworotko
The first sorbent with high CO
2
selectivity and poor water affinity addresses need for trace CO
2
remediation in confined spaces.
Christelle Vancutsem, Frédéric Achard, Jean‐François Pekel, Ghislain Vieilledent, Silvia Carboni, Dario Simonetti, Javier Gallego, Luiz E. O. C. Aragão, Robert Nasi
Three decades of satellite observations provide unprecedented information about forest change trajectories in the humid tropics.
Jiabin Feng, Yongzhuo Li, Jianxing Zhang, Yuqian Tang, Hao Sun, Lin Gan, Cun‐Zheng Ning
Two-dimensional (2D) semiconductors have emerged as promising candidates for various optoelectronic devices especially electroluminescent (EL) devices. However, progress has been hampered by many challenges including metal contacts and injection, transport, and confinement of carriers due to small sizes of materials and the lack of proper double heterostructures. Here, we propose and demonstrate an alternative approach to conventional current injection devices. We take advantage of large exciton binding energies in 2D materials using impact generation of excitons through an alternating electric field, without requiring metal contacts to 2D materials. The conversion efficiency, defined as the ratio of the emitted photons to the preexisting carriers, can reach 16% at room temperature. In addition, we demonstrate the first multiwavelength 2D EL device, simultaneously operating at three wavelengths from red to near-infrared. Our approach provides an alternative to conventional current-based devices and could unleash the great potential of 2D materials for EL devices.