- ItemSimulating Speech Perception in Bilateral Cochlear Implant Users with Asymmetric Input(2020) Zukerman, Danielle; Goupell, Matthew; Jaekel, Brittany; Milvae, KristinaUnderstanding speech in noise is difficult for cochlear-implant (CI) users. One potential reason for this difficulty is asymmetrical hearing between the two ears; that is, when one ear can process sound more effectively and clearly than the other ear. Such asymmetry may impair some CI users’ ability to fuse speech signals from both ears into a single stream. One way to test this is with an alternating speech paradigm, which is an experimental simplification of speech moving from talker to talker in a rapid conversation between a group of people. Previous studies have shown CI users perform 40% worse on alternating speech listening than normal-hearing individuals. The present study aims to examine if reduced alternating speech perception is the result of asymmetrical hearing, which could cause a listener to only use their better ear when listening to alternating speech, and to miss out on much of the signal that is present in the poorer ear. Six young normal-hearing participants were tested using a CI simulation with varying levels of signal degradation to simulate both asymmetrical and symmetrical hearing. The hypothesis was that participants will show selective attention to the ear with the clearer, less degraded signal in asymmetrical hearing conditions, and will overall perform worse in this condition compared to the symmetrical hearing condition. The results comparing the “better ear” and the asymmetric condition suggest that there is no evidence of selective attention; therefore we can reject the hypothesis. Future directions include increasing asymmetry across ears by simulating more drastic degradation in the “poorer ear”. Speech perception in noise is one of the most common issues CI users face, and quantifying the contributions of asymmetrical hearing to this problem is important for resolving this issue.
- ItemRecognition of Aminated Guests by Acyclic Cucurbiturils in Biological Conditions(2020) Shah, Rohan; Isaacs, Lyle; Zebaze, SandraThe acyclic cucurbituril Motor2 has already been well documented in its binding to several types of molecular guests in phosphate buffer. However, while these tests provide a rough idea of motor2 affinity to different types of guests, they are incomplete in that they do not reflect how motor2 actually binds in body conditions. The human body contains many proteins and macromolecules that can affect the host-guest interactions of motor2, so it is important for new binding constants to be measured for motor2 in body conditions. In order to do this, Isothermal Titration Calorimetry (ITC) was used to measure motor2 binding constants to several different guest types in several different solutions, including albumin and fetal bovine serum. It was found that when tested with cyclic, monoaminated guests, motor2 binding affinity did not decrease significantly from phosphate to protein serum solvents. This retained affinity held across several different ring sizes and shapes. Motor2 binding affinity did suffer greatly in protein serum for guests that were linear, regardless of how many amines they had. The results also indicated that more hydrophobic guests do not bind as well to motor2 once albumin and other proteins ae introduced to solution, while hydrophilic, polar guests have better affinity retention. The ITC testing results indicated that motor2 binding in body conditions is heavily dependent on the shape of the guests it is binding to, and that motor2 would be most effective at its purpose in the human body if it was used to target cyclic amines and similar types.
- ItemPredicting facial movement using electromyography and machine learning(2020) Choi, Theresa; Brustad, Abby; Morales, Santiago; Fox, NathanVideo coding participants’ behavior is inherently a subjective and time-consuming process. The purpose of this study is to support traditional video coding methods of facial expressions by using machine learning on available electromyographic (EMG) data. For this, we tested the accuracy across four machine learning algorithms (i.e., decision tree, K-nearest neighbors (KNN), multilayer perceptron (MLP), and linear support vector classifier (SVC)). Specifically, we tested their accuracy in distinguishing between (a) any facial activity versus no movement, and (b) different facial expressions (Fearful, Happy, Neutral). Success was measured by final accuracy on a pre-chosen test set. Results showed that the decision tree and KNN classifiers had the highest potential for detecting facial activity with a test accuracy of 94%. However, after plotting their decision boundaries, both had a risk of overfitting, suggesting that the best classifier could instead be a safer choice of the MLP or SVC algorithms with 84% accuracy. For classifying different facial expressions, the MLP algorithm had the highest accuracy with 88% accuracy. Overall, the conclusion is that with further development, machine learning models could simplify the video coding process. While there were some models with very high accuracies (above 90%), they tended to risk overfitting and not generalize to larger datasets. Thus, the best use of these models would be in tandem with other coding methods, such as by quickly verifying low-accuracy classifications via video coding or by outputting cutoff parameters that can be used to facilitate other analyses.
- ItemAnalysis of RNA Concentration of Influenza and other Respiratory Viruses from Dormitory Air Sampling(2020) Khan, Razeen; Bueno de Mesquita, Jacob; Milton, DonaldMaking accurate exposure assessments of airborne disease transmission is an integral part of a proactive response to outbreak events and can help track the pathway of transmission. This study aimed to assess the use of the rebreathed air equation in a dorm room setting and compare the expected exposure provided by the equation with actual viral collection determined by sampling. The study involved quantifying viral aerosol levels in the dormitory rooms of college students infected with influenza and other respiratory infections. NIOSH bioaerosol samplers collected dorm room air overnight and the viral concentration from these samples were compared against the calculated exposure value provided by the rebreathed air equation informed by direct measurements of viral shedding rates from the infected dorm residents ascertained by the Gesundheit-II bioaerosol collector. This comparison was facilitated by the rebreathed-air equation. Air samples were collected from the dormitories of nineteen participants and three participants had influenza. No virus was detected in the NIOSH samples. Data obtained from GII collection on viral shedding was then used in the application of the rebreathed air equation to predict exposure and assess how close the estimate of viral particles was to the actual results. By sampling in the dormitories of students with acute respiratory infections, we can make exposure assessments for roommates of infected students and others living in the dorms with greater accuracy by comparing actual outcomes with theoretical estimates. This work also helps improve understanding of airborne pathogen transmission in dorms and other indoor environments. The outcome of this project and future research like this helps evaluate the use of the rebreathed-air equation in predicting exposure and transmission risk under the assumption of well-mixed air.
- ItemReforestation in Peru: Effects of Mercury Contamination(2020) AYEBAE, JADA-MERCY; Andrade, Natasha; Rodriguez, MariaJada-Mercy Ayebae1*, Maria Rodriguez1, Dr. Natasha Andrade *Civil and Environmental Engineering Department, University of Maryland, College Park, MD 20742 In the Amazon Rainforest, illegal gold mining leaves water, soil and vegetation contaminated with mercury. This affects the environment and the populations dependent on the land for survival. Experiments to assess the toxicity of mercury on indigenous plant will be performed by UMD researchers next year in the Amazon. Radish and tomato seeds were used to design experiments with the purpose of understanding the effects of mercury on vegetation, specifically during germination. It was hypothesized that the control seeds (non-exposed to mercury) would grow longer and healthier than the others, to be implicated by root elongation. Seeds were treated with various concentrations of mercury (2ppm, 1ppm, 0.5ppm, 0ppm). The results showed that the seeds from the control group and those exposed to 0.5 ppm of mercury had greater root elongation than those exposed to 1 and 2ppm, after three days of germination in radishes. The tomatoes were tested in a weeklong experiment finding that, the control group had a significantly higher root length followed by 0.5 and 2ppm. These results showed that 2ppm (the greatest concentration) stunted root elongation the most in radishes and 1ppm stunted root elongation the most in tomatoes. These results support some of the hypothesis and the worries about mercury deposits in the Amazon rainforest.