93887 research outputs found
Sort by
Mhealth Hyperspectral Learning for Instantaneous Spatiospectral Imaging of Hemodynamics
Get PDFHyperspectral imaging acquires data in both the spatial and frequency domains to offer abundant physical or biological information. However, conventional hyperspectral imaging has intrinsic limitations of bulky instruments, slow data acquisition rate, and spatiospectral trade-off. Here we introduce hyperspectral learning for snapshot hyperspectral imaging in which sampled hyperspectral data in a small subarea are incorporated into a learning algorithm to recover the hypercube. Hyperspectral learning exploits the idea that a photograph is more than merely a picture and contains detailed spectral information. A small sampling of hyperspectral data enables spectrally informed learning to recover a hypercube from a red–green–blue (RGB) image without complete hyperspectral measurements. Hyperspectral learning is capable of recovering full spectroscopic resolution in the hypercube, comparable to high spectral resolutions of scientific spectrometers. Hyperspectral learning also enables ultrafast dynamic imaging, leveraging ultraslow video recording in an off-the-shelf smartphone, given that a video comprises a time series of multiple RGB images. To demonstrate its versatility, an experimental model of vascular development is used to extract hemodynamic parameters via statistical and deep learning approaches. Subsequently, the hemodynamics of peripheral microcirculation is assessed at an ultrafast temporal resolution up to a millisecond, using a conventional smartphone camera. This spectrally informed learning method is analogous to compressed sensing; however, it further allows for reliable hypercube recovery and key feature extractions with a transparent learning algorithm. This learning-powered snapshot hyperspectral imaging method yields high spectral and temporal resolutions and eliminates the spatiospectral trade-off, offering simple hardware requirements and potential applications of various machine learning techniques
Method for Estimating Pulsatile Wall Shear Stress from One-dimensional Velocity Waveforms
Get PDFWall shear stress (WSS)—a key regulator of endothelial function—is commonly estimated in vivo using simplified mathematical models based on Poiseuille\u27s flow, assuming a quasi-steady parabolic velocity distribution, despite evidence that more rapidly time-varying, pulsatile blood flow during each cardiac cycle modulates flow-mediated dilation (FMD) in large arteries of healthy subjects. More exact and accurate models based on the well-established Womersley solution for rapidly changing blood flow have not been adopted clinically, potentially because the Womersley solution relies on the local pressure gradient, which is difficult to measure non-invasively. We have developed an open-source method for automatic reconstruction of unsteady, Womersley-derived velocity profiles, and WSS in conduit arteries. The proposed method (available online at https://doi.org/10.5281/zenodo.7576408) requires only the time-averaged diameter of the vessel and time-varying velocity data available from non-invasive imaging such as Doppler ultrasound. Validation of the method with subject-specific computational fluid dynamics and application to synthetic velocity waveforms in the common carotid, brachial, and femoral arteries reveals that the Poiseuille solution underestimates peak WSS 38.5%–55.1% during the acceleration and deceleration phases of systole and underestimates or neglects retrograde WSS. Following evidence that oscillatory shear significantly augments vasodilator production, it is plausible that mischaracterization of the shear stimulus by assuming parabolic flow leads to systematic underestimates of important biological effects of time-varying blood velocity in conduit arteries
Dynamic Obstacle Avoidance for USVs Using Cross-Domain Deep Reinforcement Learning and Neural Network Model Predictive Controller
Get PDFThis work presents a framework that allows Unmanned Surface Vehicles (USVs) to avoid dynamic obstacles through initial training on an Unmanned Ground Vehicle (UGV) and cross-domain retraining on a USV. This is achieved by integrating a Deep Reinforcement Learning (DRL) agent that generates high-level control commands and leveraging a neural network based model predictive controller (NN-MPC) to reach target waypoints and reject disturbances. A Deep Q Network (DQN) utilized in this framework is trained in a ground environment using a Turtlebot robot and retrained in a water environment using the BREAM USV in the Gazebo simulator to avoid dynamic obstacles. The network is then validated in both simulation and real-world tests. The cross-domain learning largely decreases the training time (28% role= presentation style= box-sizing: border-box; max-height: none; display: inline; line-height: normal; font-size: 13.2px; overflow-wrap: normal; text-wrap: nowrap; float: none; direction: ltr; max-width: none; min-width: 0px; min-height: 0px; border: 0px; padding: 0px; margin: 0px; color: rgb(34, 34, 34); font-family: Arial, Arial, Helvetica, sans-serif; position: relative; \u3e28%28%) and increases the obstacle avoidance performance (70 more reward points) compared to pure water domain training. This methodology shows that it is possible to leverage the data-rich and accessible ground environments to train DRL agent in data-poor and difficult-to-access marine environments. This will allow rapid and iterative agent development without further training due to the change in environment or vehicle dynamics
The Role of Region and Religious Tradition in Predicting Individuals’ Expressions of Faith in the Workplace
Get PDFWhile many variables might influence an individual’s willingness to express their faith in the workplace, the role of regional context has not been fully considered. The different geographical regions in the U.S. consist of unique demographics and cultures that could shape an individual’s expression of faith at work. Moreover, these regional effects might be moderated by an individual’s specific religious tradition. Using data from a survey of U.S. adults featuring oversamples of Jewish and Muslim individuals, we utilize two unique measures of religious expression—displaying/wearing religious items at work and talking about religion at work—to assess the roles of region and religious tradition in expression of faith at work. We find that regional cultures can sometimes override religious subcultures to determine if and how people express their religion in the workplace. We find that evangelical-conservative Christians are more likely than those following most other religious traditions to say that they talk about their faith at work, regardless of the region in which they reside. However, we also find that individuals in the South tend to be more likely to express their faith in the workplace independent of their religious tradition while evangelicals in the Northwest are less so. The findings have broader implications for subcultures related to religious pluralism in an increasingly diverse U.S. society
Sources of Misinterpretation in the Input and Their Implications for Language Intervention With English-Speaking Children
Get PDFPurpose: In English and related languages, many preschool-age children with developmental language disorder (DLD) have difficulties using tense and agreement consistently. In this review article, we discuss two potential input-related sources of this difficulty and offer several possible strategies aimed at circumventing input obstacles. Method: We review a series of studies from English, supplemented by evidence from computational modeling and studies of other languages. Collectively, the studies show that instances of failures to express tense and agreement in DLD resemble portions of larger sentences in everyday input in which tense and agreement marking is appropriately absent. Furthermore, experimental studies show that children\u27s use of tense and agreement can be swayed by manipulating details in fully grammatical input sentences. Results: The available evidence points to two particular sources of input that may contribute to tense and agreement inconsistency. One source is the appearance of subject + nonfinite verb sequences that appear in auxiliary-fronted questions (e.g., Is [the girl running]? Does [the boy like popcorn]?) and as dependent clauses in more complex sentences (e.g., Help [her wash the dishes]; We saw [the frog hopping]). The other source is the frequent appearance of bare stems in the input, whether nonfinite (e.g., go in Make him go fast) or finite (e.g., go in I go, you go). Conclusions: Although the likely sources of input are a natural part of the language that all children hear, procedures that alter the distribution of this input might be used in the early stages of intervention. Subsequent steps can incorporate more explicit comprehension and production techniques. A variety of suggestions are offered
Integrating Transformative Technologies in Indiana’s Transportation Operations
Get PDFNew and emerging transportation technologies, driven by automation, connectivity, and electrification, could potentially help address the transportation sector’s persistent and pervasive problems, including those associated with safety, mobility, and energy use. For this reason, the state of Indiana, uniquely positioned to serve interstate truck traffic, sought ways to identify and incorporate these new technologies on Indiana’s highways. This report addressed the challenges and opportunities regarding the integration of transformative technologies in Indiana’s truck operations, with a particular focus on truck platooning. This report started with a review of current literature about disruptive technologies in general and truck platooning specifically. This included published information on the impacts of platooning on transportation outcomes—energy use, mobility, safety, truck operators’ comfort, infrastructure condition/longevity, emissions, and other impacts. Regarding these impacts, this report presents existing simulation models for analyzing/evaluating truck platooning. Driver comfort, in terms of the platoon inter-truck headways, was investigated using a driving simulation study in the Center for Connected and Automated Center (CCAT) human factors laboratory at Purdue. This report also identified and discussed opportunities and challenges to truck platooning, and a process was developed to identify truck-platooning sections and a multi-criteria framework for ex poste or ex ante evaluation of platooning segments. Finally, the report discusses the future trends of freight transportation in Indiana (including challenges and opportunities) in truck platooning policy and development
Full-Scale Testing of Power Transfer Roadways
Get PDFA dynamic wireless power transfer roadway refers to the integration of wireless power transfer technology into a new and existing road infrastructure to provide motive power, battery charging, or both to electric vehicles. The objective of this presentation is to evaluate the mechanical and thermal performance of the system when incorporated in both flexible and rigid pavement structures
