Machine learning concepts, concerns and opportunities for a pediatric radiologist.

Machine learning, a subfield of artificial intelligence, is a rapidly evolving technology that offers great potential for expanding the quality and value of pediatric radiology. We describe specific types of learning, including supervised, unsupervised and semisupervised. Subsequently, we illustrate two core concepts for the reader: data partitioning and under/overfitting. We also provide an expanded discussion of the challenges of implementing machine learning in children's imaging. These include the requirement for very large data sets, the need to accurately label these images with a relatively small number of pediatric imagers, technical and regulatory hurdles, as well as the opaque character of convolution neural networks. We review machine learning cases in radiology including detection, classification and segmentation. Last, three pediatric radiologists from the Society for Pediatric Radiology Quality and Safety Committee share perspectives for potential areas of development.

Controlled synthesis and transfer of large-area WS2 sheets: from single layer to few layers.

The isolation of few-layered transition metal dichalcogenides has mainly been performed by mechanical and chemical exfoliation with very low yields. In this account, a controlled thermal reduction-sulfurization method is used to synthesize large-area (~1 cm(2)) WS2 sheets with thicknesses ranging from monolayers to a few layers. During synthesis, WOx thin films are first deposited on Si/SiO2 substrates, which are then sulfurized (under vacuum) at high temperatures (750-950 °C). An efficient route to transfer the synthesized WS2 films onto different substrates such as quartz and transmission electron microscopy (TEM) grids has been satisfactorily developed using concentrated HF. Samples with different thicknesses have been analyzed by Raman spectroscopy and TEM, and their photoluminescence properties have been evaluated. We demonstrated the presence of single-, bi-, and few-layered WS2 on as-grown samples. It is well known that the electronic structure of these materials is very sensitive to the number of layers, ranging from indirect band gap semiconductor in the bulk phase to direct band gap semiconductor in monolayers. This method has also proved successful in the synthesis of heterogeneous systems of MoS2 and WS2 layers, thus shedding light on the controlled production of heterolayered devices from transition metal chalcogenides.