Master of Science in Computer Science Theses and Plan B Project Papers

Persistent link for this collectionhttps://hdl.handle.net/11299/275124

This collection contains some of the final works (theses and Plan B project papers) produced by master's degree students in the Master of Science in Computer Science graduate program. Students in this program complete either a Plan A (thesis-based) program or a Plan B (project-based) program. Additional Plan As (theses) can be found in the University of Minnesota Twin Cities Dissertations and Theses collection.

Search within Master of Science in Computer Science Theses and Plan B Project Papers

Browse

Now showing 1 - 5 of 5

Remote monitoring of patient-caregiver dyadic interactions in the hospital environment: neonatal care case study
(2025-05) Yarlagadda, Harika
Reports show that in 2020 approximately 3.6 million live births were recorded in the United States, with an estimated 9–13% of these infants requiring neonatal intensive care unit (NICU) admission due to complex medical needs. In the NICU, the parental involvement is key to helping infants recover from medical trauma. One way to help parent-infant recover from a negative experience is by introducing Kangaroo care (KC) during their NICU stay, which involves skin-to-skin contact to help strengthen the bond between parents and infants. Past research has demonstrated the health benefits of KC for both children and parents such as infant health improvement and reduced parental stress. However, resistance from healthcare workers, lack of time from healthcare workers, difficulty in coordinating schedules between parents and needed healthcare workers, infrastructure and equipment issues are some of the barriers identified that prevent KC practice. There is a need to identify alternative healthcare monitoring systems that can be used efficiently to support the interactions between the parent-child dyad during their stay in the hospital, specifically the NICU. This thesis seeks to address the critical gap by developing a HIPPA-compliant, remote healthcare monitoring system to support positive interactions between patients and caregivers. We conducted a proof of concept experiment, utilizing video recording and sensor data collection to test the feasibility of the system with adult participants.Our study evaluated the usability and effectiveness of a health monitoring system in a simulated NICU environment, involving 30 adult participants. Quantitatively, the system was rated highly for usability, with most participants finding it easy to use and reliable. Average satisfaction levels were positive, particularly in terms of system interaction and effectiveness in facilitating parental bonding, with average scores above 8 out of 10. Qualitatively, participants appreciated the technical capabilities and adaptability of the system. Specifically participants mentioned its non-intrusive monitoring and automated features enhanced caregiving. Concerns were minimal but focused on privacy due to continuous monitoring. Overall, the system demonstrated potential to improve parent-infant interactions and reduce caregiver stress in NICU settings.
Robot-Assisted Physical Therapy for cognition and motor skill enhancement in people with dementia.
(2024-08) Hassan, Isra
Dementia is a prevalent neurocognitive disorder that impacts cognition in early stages, resulting in memory loss and gradually affecting mobility and movements in carrying out daily living activities in the elderly population. It poses significant challenges in healthcare, particularly in the context of caregiving and therapeutic interventions aimed at enhancing patients’ physical and cognitive functions. This study investigates the efficacy of robot-assisted therapies to improve cognition and motor skills for individuals affected with dementia.The primary objective of this research is to evaluate whether robot-assisted therapies, facilitated by the NAO robot, can improve physical functioning, motor skills, cognitive skills, and overall well-being among dementia patients. We used NAO robot to perform 3 sets of physical exercises over a period of 10 days on participants affected with dementia and collected their Electrodermal Activity (EDA), Heart Rate, and Accelerometer data using wearable sensors during each session. We also conducted pre and post assessments to measure improvement in cognition and mobility before and after the study. Thorough statistical analysis was done on data collected during the study to identify outcomes. Results give us evidence of positive correlation between Robot-Assisted Physical Therapies and Cognition. Further analysis also show improvement in physical health and mobility after a 10 day intervention. Findings from this research have the potential to inform decision-making processes regarding the integration of robotics into dementia care, ultimately striving towards enhancing the quality of life for individuals affected by dementia.
Exploring Augmented Reality for a One-Dimensional Motion Physics Laboratory Experiment
(2023-05) Joyal, Matthew
General Physics laboratory experiments can be time-consuming to set up for each class.Additionally, students often have a specific room and limited time to conduct the experiment and draw conclusions. One advantage of moving the labs to an augmented reality setting is that setup time can be lowered. The main motivation, however, would be to allow students more time to explore the experiments on their own outside of class. Because the laboratory experiment is a computer simulation, it could provide students with options not available in a physical lab, for example, one could alter the acceleration due to gravity. With added capabilities, augmented reality laboratory experiments could be used either as a replacement for, or an extension of, traditional laboratory experiments. This paper explores the feasibility of using augmented reality to replace Physics laboratory experiments. A one-dimensional motion Physics laboratory experiment was recreated in the Unity game engine and deployed to the Microsoft Hololens 2. Study participants worked through instructions which had them collect data and perform calculations based on that data. The experiment was conducted as an exploratory study to get feedback on the user interface and how well participants were able to complete the tasks, and to gauge the general response to using augmented reality for a Physics laboratory experiment. Based on early findings, an augmented Physics laboratory experiment appears to provide a suitable replacement, but the provided material and available toolset contribute to the effectiveness of the experience.
Sharing the Load - Offloading Processing and Improving Emotion Classification for the SoftBank Robot Pepper""
(2021-04) Savela, Shawn
Pepper is a humanoid robot created by SoftBank Robotics that was designed and built with the purpose of being used for robot-human interaction. There is an application interface that allows development of custom interactive programs as well as a number of built-in applications that can be extended and used when creating other custom programs for the robot. Among the pre-installed applications are applications that will classify a person's emotion and mood using data from several data points including facial characteristics and vocal pitch and tone. Due to the Covid-19 pandemic many people have been wearing face masks in both public and private areas. Detecting emotions based on facial recognition and voice tone analysis may not be as accurate when a person is wearing a mask. An alternative method that can be used to classify emotion is to analyze the actual words that are spoken by a person. However, this feature is not currently available on Pepper. In this study we describe a software solution that will allow Pepper to perform sentiment classification based on spoken words using a neural network. We will describe the testing procedure that was used to interview participants by Pepper and compare the F1 score of each classification method with each other. Pepper was able to be programmed to use a neural network for emotion classification. A total of 32 participants were interviewed, with the NLP spoken-word analysis classification achieving an averaged F1 score of .2860 as compared to the built-in software average F1 scores of .2362 from the mood application, .1986 from the vocal tone and pitch application, and .0811 from the facial characteristics application.
Estimating File Compressibility Using File Extensions
(2021-07) Powers, Carson
Adaptive compression systems dynamically choose a compression strategy — including no compression — by monitoring CPU usage, output rate, expected time to compress, and perhaps most importantly, the estimated compressibility of the data. Many adaptive compression systems were designed with the assumption that files with the same filename extension will compress roughly to the mean compression ratio (the ratio of compressed size to original size) of some set of files with the same extension. This implies that the compression ratio distribution follows a normal distribution. Though a normal distribution of compression ratios may seem intuitive, this assumption lacks strong empirical supporting evidence. To test this assumption, we built a tool to compress real-world files from many participants, storing the compressed size, original size, file extension, and other metadata. The results of three tests for normality indicate that none of the file extensions we analyzed have a normal distribution, though for some extensions, not all three tests agree. Furthermore, quantitative analysis reveals that files with the same extension compress according to multiple different distributions, and we identified some readily accessible metadata that can separate these files into simpler distributions. We conclude with a discussion of the utility of mean compressibility as an estimator and the implications this study has for future research in adaptive compression.

University Digital Conservancy

University of Minnesota Duluth

Browse

Recent Submissions