More about HKUST
Deep Learning-Driven Sensing Systems for Chronic Disease Management: A Survey
PhD Qualifying Examination Title: "Deep Learning-Driven Sensing Systems for Chronic Disease Management: A Survey" by Mr. Chi XU Abstract: Chronic diseases represent a significant public health challenge, contributing to approximately 40 million deaths annually and accounting for more than 70% of global mortality, as reported by the World Health Organization (WHO). However, effective management of these diseases remains a critical issue worldwide, necessitating innovative solutions. Fortunately, the emergence of deep learning and various sensing techniques has led to the development of powerful systems that demonstrate substantial potential in chronic disease management. Due to their capacity to learn from vast amounts of data, identify complex patterns, and make informed decisions that mimic human cognitive functions, neural networks are able to excel in tasks such as image recognition, natural language processing, and predictive analytics. Additionally, different healthcare scenarios present unique requirements, indicating that tailored sensors and modalities should be employed to address specific needs effectively. To achieve better performance in this domain, combinations of deep learning and various sensors are essential for optimizing treatment strategies and enhancing patient outcomes. This survey explores the transformative potential of deep learning-driven sensing systems in chronic disease management, emphasizing their ability to analyze vast amounts of heterogeneous data for accurate predictions and real-time monitoring. We first provide an overview of fundamental concepts and recent advancements in deep learning and sensing systems. Next, we will give a detailed literature review on specific applications across various chronic diseases, including respiratory, cardiovascular, and neurological disorders. Furthermore, we discuss our research efforts in respiratory disease assessment and intervention. In conclusion, we identify critical future directions for enhancing chronic disease management systems. This review aims to highlight the significant impact of deep learning and sensing technologies on improving patient healthcare outcomes. Date: Monday, 10 February 2025 Time: 1:00pm - 3:00pm Venue: Room 4472 Lifts 25/26 Committee Members: Prof. Qian Zhang (Supervisor) Dr. Xiaojuan Ma (Chairperson) Prof. Song Guo Dr. Xiaomin Ouyang