Program

Abstract: " Much work has been done in recent years advance technology and standards in Prognostics and Health Management. This work has addressed issues in mechanical, hydraulic, and structural PHM, and more recently, considerable attention is being given to electronic PHM as well. Even so, the majority of work that has been done has been in the context of “on-platform” PHM, focusing on sensors, data acquisition, data processing, modeling, and prediction. In the commercial and military aerospace community, however, additional information about the health of a system is collected “off-platform” through general and specialized support equipment, called automatic test equipment. In this talk, I provide an overview of work going on with the Department of Defense in terms of the creation and management of an Automatic Test Systems Framework and describe a current initiative underway developing a new IEEE standard for PHM and ATS. I also introduce work being done through support from the DoD to develop tools to support PHM in the context of the ATS Framework. "

Abstract: " In recent years, the Internet of Medical Things (IoMT) has played a significant role in the healthcare industry. It has not only empowered medical services to handle complex real-time functions but has also greatly enhanced patient monitoring, benefitting patient care and medical research. Hence, there is a need to ensure that the data received from various sensors stay secure. Considering these advancements in the healthcare industry, concerns have arisen regarding the protection of sensitive medical data within the IoMT framework. Blockchain offers transparent, immutable, and decentralized data storage and its versatility holds the potential to reshape the IoMT framework. It can facilitate personalized and secure patient data management, revolutionize traditional healthcare practices, establish robust data-sharing mechanisms, enhance pharmaceutical supply chain management’s efficiency, and enable the traceability of drugs. Integrating IoMT and blockchain technologies holds significant promise for maximizing their advantages within the healthcare sector, ranging from enhanced data security and integrity to streamlined interoperability and transparency. This integration fortifies the robustness of healthcare data management. It facilitates the seamless sharing and traceability of medical information, ultimately improving patient care and heightened trust within the healthcare ecosystem. To bridge the gap between theory and practice, the BlockMedHealth framework is proposed—a lightweight, decentralized real-time monitoring and prescription management system. This framework utilizes IoMT and blockchain technology to store Electronic Health Records (EHRs) securely, ensuring data integrity, anonymity, and interoperability. Performance evaluations demonstrate BlockMedHealth's effectiveness in processing and storing Remote Patient Monitoring (RPM) data. In addition to that, a framework based on a secure consortium blockchain that prioritizes data privacy and employs time-based authentication to streamline patient data monitoring. The findings of this research work indicate that the proposed solution effectively enhances the management of IoMT data by providing improved privacy and security. Blockchain technology is also transforming supply chain security, in addition to the healthcare industry. In another related research, we explore two blockchain-based methodologies to address privacy, traceability, and counterfeiting within the supply chain sector. The first methodology employs a permissioned blockchain with a Centralized Data Store (CDS) whereas the second methodology introduces a comprehensive blockchain system with Proof of Elapsed Time (PoET) consensus and advanced privacy algorithms, successfully addressing privacy concerns and safeguarding against counterfeiting. From securing medical data to combating counterfeits in the supply chain, these innovations promise a future where transparency, security, and efficiency converge for the betterment of industries and the well-being of individuals. "

Abstract: " Healthcare in the 21st century will be transformed by various digital revolutions. Of particular importance will be four revolutions: 1) Artificial Intelligence/Machine Learning (AI/ML); 2) Omics; 3) Virtual Reality/Augmented Reality (VR/AR); and the 4) Internet. Artificial Intelligence (AI) is concerned with the development of computer programs that emulate the intelligence of humans, i.e., AI is deeply concerned with the understanding of human problem-solving strategies and incorporating (or simulating) these strategies into computer programs. The “Omics” paradigm can be viewed as the study of a domain in a massive scale, at different levels of abstraction, in an integrative manner. VR and AR integrate the physical and virtual worlds and provide an immersive experience for the human. The Internet, which has spanned several networks in a wide variety of domains, is having a significant impact on every aspect of our lives. The P9 medicine concept -- personalized, predictive, preventive, participatory, pervasive, precise, privacy-preserving, protective, and priced reasonably – forms the basis for a digital health framework. In this talk, I will discuss the P9 concept and its manifestation in the various digital revolutions, with a plan for an implementation strategy for digital wellness, which can be viewed as the use of information technology for ensuring robust physical and mental health. "

Abstract: " Automatically improving and repairing software using search-based methods is an active research topic. Many current systems use existing source code as the ingredients of repairs, either through evolutionary computation derived random mutation or other heuristic operators. However, these code transformation operators are not always well-matched to the granularity of the source code on which they operate. This tutorial will provide an overview of the automated program repair landscape and discuss a tool to improve automated program repair tools via a static source-to-source preprocessing step to produce code with more uniform granularity that exposes relevant program components to the repair process. Additionally the tutorial will investigate applications of automated program repair to produce more fault tolerant code utilizing the byproduct of automated program repair, diversity. "

Abstract: " There is an increased need for onboard decision-making capabilities in cyber-physical systems be it in energy, automotive, aviation, space, or other industries as they aim for increased efficiency, resiliency, and mission assurance capabilities. Emerging next-gen technologies such as multi-rover planetary missions, distributed satellites, unmanned ground and aerial vehicle operations and smart grid systems rely on in-time risk assessment and autonomous decision-making. One critical piece of the autonomy puzzle is reliable prediction of system behavior under time-varying and potentially uncertain environmental conditions. Further, if agent states change during operation such as initiation of faults or degradation, reliable diagnostic tools need to be investigated. In this tutorial, we will revise approaches that integrates existing physics-based and data-driven models of agents interacting with probability models of the environment and component operation state. Role of existing PHM methodologies as they feed into decision-making under uncertainty will be studied. Balancing critical trade-offs between high-fidelity prognostic models, prediction time-horizons and the computational requirements for in-time cost-effective decision-making will be discussed through the implementation of surrogate models. Finally, the audience will be introduced to a real-time application of in-time trajectory planning of an unmanned aerial system (UAS) based on its PHM assessments under uncertain and varying wind conditions. "

Abstract: " The successful adoption of PHM technologies and strategies will depend on our ability to communicate effectively with end users and the public, particularly around the importance of being prepared for technological failures. But communicating to non-experts about uncertainty, product safety, reliability and risk can present significant challenges, as recent public responses to climate change, pandemic risk, and other health scares have shown. In this tutorial, we explore how these failures in science communication have impacted the public’s understanding of science and their response to risk in these areas, then examine the communication strategies that led to these poor outcomes. In particular, we interrogate some prominent assumptions that are often made about non-expert audiences, specifically the idea that the public is homogenous, with similar goals, background knowledge, and purposes for learning about technology. Acknowledging this error, we introduce the concept of “Framing Science” as a corrective, which argues that in order to communicate effectively with the public, scientists and researchers need to do more than learn techniques for “dumbing-down” or “translating" their research while maintaining accuracy: they need to find ways to make complex topics more personally relevant to end users. Without these connections, non-expert audiences are not compelled to engage with and understand technical information, nor to act on the advice of scientific experts. We conclude by laying out key strategies for: (1) understanding an audience’s needs and background; (2) finding relevant approaches to engage non-experts in topics relevant to Prognostics and Health Management; and finally, (3) deploying the right rhetorical techniques (eg. narrative, metaphor, etc.) to ensure that the public understands how PHM research can better manage their relationship to medical and other technologies. "

Abstract: " The Digital Twin (DT) is a highly promising technology within the digital transformation realm. It is a virtual replica of a physical system that mirrors its behavior. The emulation forecasts potential breakdowns and openings for change, offering real-time recommendations and optimizing or managing unexpected occurrences. Crafting this virtual model involves a complex process reliant on substantial data and precise models that accurately mirror reality. Currently, this modeling approach varies based on the specific use case, necessitating a tailored solution for virtual reality modeling within the digital twin framework. A reference methodology/approach for DT design and implementation in Prognostics and Health Management Systems will be presented, centered around the concept of "modeling patterns" and their inherent consistency. These patterns, conceived as data-driven, can autonomously derive from data through a specialized approach aimed at achieving a consistent trait. This enables their application and reuse across modeling scenarios or problems of varying degrees of similarity. The efficacy of these consistent modeling patterns is showcased through real industrial case studies where a dedicated DT was constructed utilizing the proposed approach Prognostics and Health Management reasons. "

Abstract: " The landscape of Personalized Medicine (PM) stands at a critical crossroads, marked by both unprecedented high-risk challenges and transformative high-reward opportunities, stemming in part from recent breakthroughs in Artificial Intelligence (AI). Unlike the traditional approach of focusing solely on tumor features, PM targets tailoring treatments based on tumor, environmental, lifestyle, and/or patient molecular profiles. The ultimate objective of PM is reshaping diagnosis/prognosis and treatment effectiveness prediction for development of reliable therapies. In this defining stage of PM evolution, on the one hand, identification, detection, and recognition of biomarkers (at the molecular or cellular levels) play a crucial role. On the other hand, due to the diversity of patient profiles, reliable prediction of individual responses to medications represents a major hurdle in PM. A paradigm shift has, therefore, swept through the medical community, moving from traditional disease-focused approaches to data-centric and computational disease identification. Deep Neural Networks, in particular, Foundation Models (FMs) are posed to become pivotal in recognizing disease patterns, tailoring treatment plans, and enhancing prognosis accuracy. Unlike conventional DNNs, FMs can delve into patient-disease correlations, positioning themselves as an indispensable and integral part of the PM concept. Our discourse begins by exploring reliable radiomics feature extraction using DL and FMs, illustrating how these computational models reliably learn latent patterns and textures from medical images, translating them into quantifiable insights. The effectiveness of this approach in various medical applications, ranging from oncology to neurology, will be showcased, emphasizing its potential in revolutionizing patient prognosis and treatment planning. "

Abstract: " Reinforcement learning is a critical paradigm in artificial intelligence that enables an agent to learn optimal behaviors through trial-and-error interactions with environments. It exhibits significant potential for solving complex decision-making problems in engineering scenarios and has been regarded as a transformative approach within the reliability and maintenance optimization community. This talk will introduce the theoretical foundations of reinforcement learning, as well as its applicability and advancements in reliability and maintenance optimization. It will provide a comprehensive tutorial on applying reinforcement learning to address reliability and maintenance optimization problems, including Markov decision processes modeling, design and implementation of reinforcement learning algorithms, and case studies of engineering instances. Additionally, challenges and future directions associated with the application of reinforcement learning in reliability and maintenance optimization will be discussed. "