Congratulations to Rafal Krzysiak for successful Ph.D. dissertation defense.
On 11/17/2025 Monday 9:00-11:30AM, Rafal Krzysiak successfully defended his Ph.D. dissertation winning unanimous pass from his committee members. Rafal's thesis title is: "Smart Health Monitoring Using Explainable AI - From Human Physiology to Earth Remote Sensing". Congratulations!
Abstract:
Advancements in Artificial Intelligence (AI) and Explainable AI (XAI) have revolutionized applications across healthcare and planetary science. This dissertation proposal explores the integration of XAI to address critical challenges in smart health monitoring and Earth remote sensing. In healthcare, the research introduces frameworks leveraging wearable devices for continuous physiological monitoring and stress prediction, enhancing transparency in AI-driven diagnostic tools. It proposes the XCardio-Twin, a digital twin framework for explainable cardiovascular monitoring, bridging clinical insights with advanced computational models. For planetary science, this work develops innovative methodologies for land-use classification and methane emission quantification using hyperspectral satellite data. It presents a novel fractional-order calculus-based framework for methane concentration retrieval, enabling precise identification and quantification of methane emissions. The research also introduces the XDT-Wildfire framework, which models wildfire spread dynamics with XAI-enhanced predictive capabilities, integrating multi-scale data sources such as satellite, drone, and ground-level observations. Additionally, the dissertation details the development of a Level-II Digital Twin for the International Space Station’s tunable laser spectrometer. This behavior matched framework enhances prediction and system optimization under variable conditions in microgravity environments, advancing thermal and structural modeling of spaceborne instruments. By combining XAI methodologies with cutting-edge technologies, this dissertation advances the fields of personalized healthcare and planetary health. These contributions demonstrate how transparent AI frameworks can provide actionable insights, foster trust, and address complex challenges in high-stakes applications across diverse domains.
Created and last updated 11/18/2025 by Prof. YangQuan Chen
