Health Scholar Showcase
Permanent URI for this collectionhttps://hdl.handle.net/10877/16500
The Health Scholar Showcase is an annual event hosted by Texas State University’s Translational Health Research Center, which seeks to improve health by connecting faculty and community partners to engage in innovative research. Health Scholar Showcase highlights some of the best health research happening on campus.
Learn more about Health Scholar Showcase: https://healthresearch.txst.edu/events/health-scholar-showcase.html
Browse
Browsing Health Scholar Showcase by Department "Computer Science"
Now showing 1 - 8 of 8
Results Per Page
Sort Options
Item AI-Powered Auxiliary Medical Diagnostic Systems(2024-03) Farias, Mylène C. Q.Deep Learning models are being used to analyze medical data and, most specifically, medical images, and to identify patterns and abnormalities that may not be (YET) visible to radiologists and physicians in general. These auxiliary diagnostic systems allow for an early detection of chronic diseases, such as heart conditions and cancer. AI models can process large amounts of data quickly and accurately. They can also be used to track health data over time and identify suspicious changes. Finally, AI models can be used to identify rare diseases and conditions that are difficult for humans to diagnose. But the area still faces several challenges: Availability of balanced datasets; Assurance of accuracy and reliability; Explainability; Privacy and security; Robustness to diversity in formats, degradations, etc.Item An Empirical Study on AI-Powered Edge Computing Architectures for Real-Time IoT Applications(2024-03) Ngu, Anne H. H.; Yasmin, AwatifEdge computing is indispensable for IoT applications, handling data from billions of devices and expected to surpass 41.6 billion installations by 2023. It facilitates swift decision-making at the device level. It conserves network bandwidth by processing data locally, making it suitable for resource-constrained or costly networks. Bolsters privacy and security by storing data locally, particularly crucial for applications that involves processing personal data.Item Controlling Epidemic Spread Under Immunization Delay Constraints(2024-03) Lee, Chul-HoNo abstract prepared.Item Dynamic and Lightweight Encryption towards Data Confidentiality in Medical loT Devices (MloT)(2023-04) Hou, Tao; Wang, TaoMedical IoT is growing at an exponential rate because of its promises to help advance patient engagement and improve delivery of health care. However, it also incurs a range of security concerns. As medical IoT devices may gather hosts’ geographical information, monitor patients’ privacy activities, and record clients’ biometric features, one of the critical concerns is how to preserve the confidentiality of such sensitive data. Specifically, because of the broadcast nature of wireless signal, the communication and data transmission with Medical IoT devices are usually vulnerable to eavesdropping attacks. Intuitively, cryptography encryption methods can be applied to encrypt all the conversation from medical IoT devices. However, as medical IoT devices, such as Implantable medical devices (IMDs), are usually featured with limited computational capacity and limited power, they may not afford expensive cryptography operations by conventional encryption methods like AES or RSA. To cope with the limited resources of medical IoT devices, we propose a novel encryption scheme, named Dynamic Wireless Channel Pad (DyWCP), that takes advantage of dynamic signal variation in wireless context to achieve the confidentiality of sensitive data.Item L3CatTXST: NIH Long COVID Computational Challenge (L3C) Results(2023-04) Tesic, Jelena; Musal, RasimNational COVID Cohort Collaborative L3C challenge: Determine if the patient who has tested positive for SARS-CoV-2 in an outpatient hospital setting (ICU or non-ICU) developing PASC/Long COVID. The N3C's data consists of existing patient records at 94 participating institutions. The data itself can only be accessed through a secure cloud portal hosted by NCATS known as the Enclave With collaborative efforts it consists of: 20 billion rows, 1,757.1 million clinical observations, 16.4 million patients, and 6,438,192 SARS-CoV cases. Under the university's DUR we have access to Level 2 data.Item Learning Loss Recovery After COVID-19 Pandemic in Texas Public Schools(2023-04) Ervin, Philip; Feng, Li; Payan, Daniel; Tesic, Jelena; Tesic, JelenaCOVID-19 school reopening decisions were difficult for policymakers since there was no consensus on the impact of school reopening on the spread of COVID-19. Learning loss was documented in many states including Texas. If we can identify most impactful factors on learning loss from publicly available data sources during pandemic, we can help policy makers make more informative decisions on learning recovery.Item P-Fall: Personalization Pipeline for Fall Detection on Wearables(2023-04) Ngu, Anne H. H.; Yasmin, Awatif; Mahmud, TarekNo abstract prepared.Item The Impact of Synthetic Data on Fall Detection Application(2024-03) Ngu, Anne H. H.; Debnath, MinakshiThe accurate recognition of the dynamic of fall using deep learning requires a lot of data. Three different methods for creating realistic synthetic fall data utilizing generative AI with diffusion, fall data extraction from 2D video recordings, and traditional data augmentation techniques are explored.