Speaker Title: Exploring the Potential of Topological Deep Learning in Advancing Medical Imaging
Dr. Yashbir (Yash) Singh, Ph.D., is an Assistant Professor of Radiology at the Mayo Clinic, Rochester, MN. His accomplished career includes a role as a Medical Scientist at WVU Medicine, USA, a DAAD fellow, and he is an active member of the New York Academy of Sciences. Dr. Singh's research concentrates on deep learning-based artificial intelligence (AI) and topological data analysis within medical imaging, specifically enhancing the interpretability of AI models and discovering new disease imaging biomarkers.
Abstract: Topological Deep Learning (TDL) has emerged as a powerful tool in medical imaging, promising a revolution in diagnostic and therapeutic strategies. This talk explores the potential of TDL in advancing medical imaging, emphasizing its role in improving human-computer interaction (HCI) in healthcare. TDL leverages the intricate geometric and topological structures underlying data, enhancing the predictive power of traditional deep learning models. The integration of TDL into healthcare presents an opportunity to optimize diagnostic accuracy, minimize human error, and offer personalized treatment plans. This talk also discusses the significance of intuitive HCI approaches in medical imaging, critical in facilitating seamless interaction between healthcare professionals and complex AI systems. Creating intelligent interfaces that enable the interpretation of TDL outputs and support clinical decision-making achieves this. The potential challenges and ethical considerations of implementing TDL in healthcare are also addressed. This talk underscores the transformative potential of TDL in medical imaging and its capability to redefine HCI, fostering a more efficient, accurate, and patient-centric approach in healthcare.
Speaker Title: Toward Human-Centered AI to Improve Decision Making in Healthcare
Hyunggu Jung is an associate professor of Computer Science & Engineering and an associate professor of Artificial Intelligence at the University of Seoul and directs the Human-Centered Artificial Intelligence Lab (HCAIL). He received his B.S., M.Math, and M.S. from KAIST, the University of Waterloo, and Stanford University, respectively, all in Computer Science. He also received his B.S. with a minor in Business Economics from KAIST and his Ph.D. in Biomedical and Health Informatics from the University of Washington School of Medicine. Further, he worked at Microsoft Research and PARC, a Xerox company as a research intern, respectively. His research interests lie at the intersection of human-centered AI, health informatics, social computing, and accessibility & aging. His research aims to advance AI research through design and engineering to support people with special needs (e.g., older adults, streamers with visual impairments, North Korean defectors with depression) across multiple domains: health, social media, sharing economy, and education. He is a recipient of the Korean Government Scholarship and Mogam Science Scholarship. For more information on his recent research activities.
Abstract: In this talk, I will present my recent research efforts in the field of human-centered artificial intelligence (AI). Over the past few years, my research has focused on the early-stage development of innovative technologies through design and engineering to support people with special needs across multiple domains: health, social media, accessibility, and sharing economy. For the rest of the talk, I will focus on describing my approaches using design methods and mathematical models to develop human-centered AI technologies. I will conclude with my future research directions, including the development of tools that leverage data by reflecting the needs and barriers of AI stakeholders (e.g., healthcare providers, managers, and patients affected by AI-based decisions), and the development of effective human-centered AI systems for improving decision making in healthcare.
Speaker Title: Exploratory Visual Analysis of Hospital Networks for Aiding Clinical Researchers
Abstract: This talk delves into the role of visual analytics in healthcare, elucidating its potential to improve patient care and research methodologies. This talk introduces a visual analytic study on hospital networks. Inter-hospital coordination, especially with the advent of endovascular thrombectomy (EVT), is vital for optimal stroke treatment outcomes. While many studies have focused on quantitative analyses of hospital networks, there's a gap in topological examinations. This talk introduces a framework that constructs and analyzes networks from stroke patient transfer data. The tool visualizes national hospital network structures, delves into detailed structures via dynamic queries, and highlights hub-and-spoke configurations within clusters. It will also discuss the future direction of the research regarding the evolution of network structures over time and explore simulations of networks based on the changing roles (e.g., hub and spoke) of hospitals.
Speaker Title: Enriching Cultural Heritage through HCI-Infused AI: ASI-Protected Temples and Monuments
Abstract: This talk delves into the symbiotic relationship between AI, deep learning, and computer vision in revitalizing India's cultural heritage, with a specific focus on ASI-protected temples and monuments. By conquering challenges in documentation, restoration, maintenance, and visitor engagement, these cutting-edge technologies are driving a resurgence in cultural preservation. The talk showcases pioneering implementations that exploit AI, deep learning, and computer vision to safeguard and propagate India's cultural legacy. It encompasses:
Speaker Title: How can healthcare services benefit from artificial intelligence?
Abstract: In this talk, we will untangle the main concepts and AI, their purposes, usage and expectations in healthcare, and study main approaches, architectures, and principles. Special focus will be given to generative AI, personalization, and social robotics in healthcare. Enriched with many current research case studies and examples, this tutorial will provide an excellent learning opportunity for computer scientists, health psychologists, health scientists and interaction designers interested in this exciting multidisciplinary field and aid in developing an articulated opinion on whether AI in healthcare a sword of Damocles is indeed.
Speaker Title: Building Inclusive Interfaces to Assist Persons with Disabilities for Effective Human Computer Interaction
Abstract: The term “Human” in the Human Computer Interaction refers to a heterogenous group of users. Inclusion is the key to effective Human Computer Interaction. In this talk, I would like to focus on a special group of users, Persons with Disabilities (PwD). The PwDs are often a marginalized group of users who face significant barriers interacting with the digital systems. In this talk, I would like to explore various interaction level barriers faced by persons with disabilities. The potential solutions to some of these barriers can be through thoughtful interface design. An interface need to be intelligent and at the same time it has to be inclusive too. Some of the barriers in the interfaces make the persons with disabilities a soft target in various cyberattacks. This talk will focus on adapting the latest developments in the field of Artificial Intelligence for the design of Intelligent and Inclusive Interfaces that aims to provide a barrier-free digital experience for persons with disabilities.
Speaker Title: Wi-Fi Sensing: Principle, Implementation and Applications for Human Activity and Gesture Recognition
Abstract: This talk is an attempt to demonstrate a device free general purpose Wi-Fi sensing system to track events and recognize activities even through the wall and other materials using the Channel State Information (CSI) values extracted from the received Wi-Fi signals at the receiver end. The received signal characteristics change with the presence of the human beings, and their activities affect the signal propagation, resulting from reflection and scattering. The activities can be recognized by analyzing the CSI values corresponding to different sub-carriers of the received signal. CSI values contain fine grain information such as amplitude and phase to achieve better sensing accuracy with a unique pattern that can be observed corresponding to each activity and material. We will present our experience of developing the transmitter and receiver hardware modules together with the necessary software for capturing the CSI from Wi-Fi signals and conducted multiple experiments using the low power, low cost ESP-32 Wi-Fi module and Intel 5300 NIC module, for human presence, activity detection, material detection, ambient condition in indoor environments.
Speaker Title: IoT-Based Unobtrusive Physical Activity Monitoring System for Predicting Dementia
Abstract: Mental health-related disorders are common in elderly populations. Among the various mental health disorders, one most significant threat is dementia, and prediction of dementia has become an important issue related to well-being in old age, because the disease progression of dementia can be slowed by early diagnosis and disease control. In this paper, we propose an unobtrusive dementia-prediction system for monitoring physical activities of elderly persons either living alone or as a couple in different house structures, achieved through passive infrared (PIR) motion sensors combined with data processing. The proposed feature extraction algorithm extracts feature values related to physical activities from simple passive infrared sensors located in each room space. We then apply a variety of common popular classification models, including Deep Neural Networks (DNNs), to predict the risk of dementia in a sensor-enabled home. We implemented and validated algorithms on data collected for over a month from 18 participants who were engaged with a variety of living conditions. The proposed system was effective in predicting dementia risk, with up to a 0.99 area under the curve (AUC) using DNN with principal component analysis (PCA) and a quantile transformer scaler. In terms of the result based on leave-one-subject-out (LOSO) analysis, an accuracy of 63.38% was achieved using DNN with PCA and a standard scaler. The proposed methodology is non-invasive and cost-effective, and can be used for a variety of long-term monitoring and early symptom detection systems, helping caregivers provide optimal interventions to elderly individuals at risk for dementia.
Speaker Title: TBA
Speaker Title: Exploring Quantum AI for Human-Computer Interaction
Abstract: This talk delves into the transformative potential of Quantum Artificial Intelligence (AI) in reshaping human-computer interaction. By harnessing the principles of quantum computing, including quantum algorithms and entanglement, we investigate how Quantum AI can overcome classical computing limitations. Through a comprehensive analysis, we envision a future where Quantum AI systems revolutionize user experiences, unlocking new frontiers in computation and problem-solving. This talk sheds light on the exciting frontier of Quantum AI, offering insights into its potential applications and implications for the future of human-computer interaction.