We have recently seen many successful applications of recurrent neural networks (RNNs) on electronic medical records (EMRs), which contain histories of patients' diagnoses, medications, and other various events, in order to predict the current and future states of patients. Despite the strong performance of RNNs, it is often challenging for users to understand why the model makes a particular prediction. Such black-box nature of RNNs can impede its wide adoption in clinical practice. Furthermore, we have no established methods to interactively leverage users' domain expertise and prior knowledge as inputs for steering the model. Therefore, our design study aims to provide a visual analytics solution to increase interpretability and interactivity of RNNs via a joint effort of medical experts, artificial intelligence scientists, and visual analytics researchers. Following the iterative design process between the experts, we design, implement, and evaluate a visual analytics tool called RetainVis, which couples a newly improved, interpretable, and interactive RNN-based model called RetainEX and visualizations for users' exploration of EMR data in the context of prediction tasks. Our study shows the effective use of RetainVis for gaining insights into how individual medical codes contribute to making risk predictions, using EMRs of patients with heart failure and cataract symptoms. Our study also demonstrates how we made substantial changes to the state-of-the-art RNN model called RETAIN in order to make use of temporal information and increase interactivity. This study will provide a useful guideline for researchers that aim to design an interpretable and interactive visual analytics tool for RNNs.
Tailoring the spectrum of thermal radiation at high temperatures is a central issue in the study of thermal radiation harnessed energy resources. Although bulk metals with periodic cavities incorporated into their surfaces provide high emissivity, they require a complicated micron metal etch, thereby precluding reliable, continuous operation. Here, we report thermally stable, highly emissive, ultrathin (<20 nm) tungsten (W) radiators that were prepared in a scalable and cost-effective route. Alumina/W/alumina multiwalled, submicron cavity arrays were fabricated sequentially using nanoimprinting lithography, thin film deposition, and calcination processes. To highlight the practical importance of high-temperature radiators, we developed a thermophotovoltaic (TPV) system equipped with fabricated W radiators and low-bandgap GaSb photovoltaic cells. The TPV system produced electric power reliably during repeated temperature cycling between 500 and 1200 K; the power density at 1200 K was fixed to be approximately 1.0 W/cm 2 . The temperature-dependent electric power was quantitatively reproduced using a one-dimensional energy conversion model. The symmetric configuration of alumina/W/alumina multiwall together with the presence of a void inside each cavity alleviated thermal stress, which was responsible for the stable TPV performance. The short-current-density (J SC ) of developed TPV system was augmented significantly by decreasing the W thickness below its skin depth. A 17 nm thick W radiator yielded a 32% enhancement in J SC compared to a 123 nm thick W radiator. Electromagnetic analysis indicated that subskin-depth W cavity arrays led to suppressed surface reflection due to the mitigated screening effect of free electrons, thereby enhancing the absorption of light within each W wall. Such optical tunneling-mediated absorption or radiation was valid for any metal material and morphology (e.g., planar or patterned).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.