A. Prof. Ata Jahangir Moshayedi
Jiangxi University of Science and Technology, China
IEEE senior member
Biography:
Dr. Ata Jahangir Moshayedi is an Associate Professor at Jiangxi University of Science and Technology in China, holding a Ph.D. in Electronics-Robotics from Savitribai Phule University, Pune, India. He is a distinguished member of IEEE as a Senior Member and ACM, alongside being a Life Member of both the Instrument Society of India and the Speed Society of India.
His contributions extend beyond academia, as he actively participates in various IEEE conferences, serving in roles such as Keynote,TPC chair, reviewer, and editorial team member for conferences and various in ternational journals.
Dr. Moshayedi's scientific achievements are substantial, with over 90 articles published in prestigious national and international journals. Alongside his research publications, he has authored three books and holds two patents and 12 copyrights, showcasing his pioneering contributions. Notably, his latest book, "Unity in Embedded System Design and Robotics: A Step-by-Step Guide," published by CRC, is recognized as the first book to integrate VR, robotics, and embedded systems.
His research interests encompass a wide range of topics, including robotics and automation, sensor modeling, bio-inspired robots, mobile robot olfaction, column tracking, embedded systems, machine vision, virtual reality, and artificial intelligence.
A. Prof. Feng Cao
Jiangxi University of Science and Technology, China
Biography:
He received his PhD in Computer Science and Technology from Southwest Jiaotong University in June 2020. At present, he is in charge of 1 provincial Education Department project and 1 school-level project, and participates in 1 National Natural Science Foundation project as a core member; Published more than 10 papers in SCI, EI and other important journals and conferences at home and abroad; Applied for 3 invention patents; International first-order logic automatic theorem prover CSE, CSE_E series developers, developed two sets of software credibility verification system based on logical reasoning. Research interests: Intelligent information processing, automatic reasoning, automatic theorem proving in first-order logic, artificial intelligence.
A. Prof. Yishan Chen
Jiangxi University of Science and Technology, China
Biography:
Yishan Chen received the Ph.D. degree from the College of Computer Science, Zhejiang University, Hangzhou, China, in 2022.,She is currently working as an Associate Professor with the School of Information Engineering, Ganzhou, Jiangxi University of Science and Technology, Ganzhou, China. Her papers have been published in some well-known conference proceedings and international journals, such as IEEE ICWS, IEEE Transactions on Mobile Computing, and MONET. Her research interests include cloud computing, service computing, edge computing, machine learning, and big data.
A. Prof. Miaomiao Liang
Jiangxi University of Science and Technology, China
Biography:
Miaomiao Liang received the Ph.D. degree in pattern recognition and intelligent systems from Xidian University, Xi'an, China, in 2018.,She is currently an Associate Professor with the School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou, China. Her research interests include computer vision, machine learning, and hyperspectral image processing.
Researcher Zeashan Khan
Interdisciplinary Research Center for Intelligent Manufacturing & Robotics, KFUPM, Saudi Arabia
IEEE senior member
Biography:
Zeashan Khan is currently associated with the intelligent manufacturing and Robotics research centre (IRC-IMR), KFUPM, Saudi Arabia. He obtained his Ph.D. in Automation and Computer Integrated Manufacturing from Grenoble Institute of Technology, France. His areas of interest include robotics, cyber-physical systems in manufacturing & process industry and healthcare. He has authored/co-authored more than 80 technical papers including two books. He is a senior member of IEEE, InstMC and ASME.
Title:AI in Research: Tools & Techniques
Abstract: In today's era of artificial intelligence (AI), researchers can leverage AI-powered tools to significantly enhance their workflow. AI can streamline processes such as prompt engineering for targeted literature searches, crafting optimal research questions, and intelligent data curation. These technologies automate repetitive tasks, improving efficiency and accuracy while enabling researchers to focus on deeper analysis and innovation. Through AI-driven literature synthesis and theory development, scholars can gain valuable insights, saving time while improving the quality of their research. This talk will explore key AI platforms that are transforming research practices, their potential limitations, and how to effectively cite these tools in academic work. By incorporating AI into the research process, professionals can gain a competitive edge, adopting innovative methods that improve both productivity and the overall quality of their research.
A. Prof. Abolfazl razi
Clemson University, USA
IEEE senior member
Biography:
Abolfazl Razi is an Associate Professor of Computer Science in the School of Computing at Clemson University. He held two postdoctoral positions in the field of Machine Learning and Bioinformatics, respectively, at Duke University (2013–14) and Case Western Reserve University (2014–15). He also served several years in the Wireless and SmartCard industry holding R&D and Management Positions.
His research area spans the interplay of AI, Machine Learning, and Secure Networking with Applications to Aerial Monitoring Systems, AI-based Networking, Remote Health Monitoring, Nano-Scaled Visual Identifiers, Driving Safety Analysis, and Zero-Trust IoT Environments. His research is supported by NSF, NIH, US Airforce, Arizona Commerce Authority, MIT Lincoln Lab, BMW Research, Philips Healthcare, and the Arizona Board of Regents.
His research results are published in 3 Book Chapters, and more than 100 Peer-reviewed Journal Articles and Conference Papers. He also has developed several software packages and product prototypes. He is the co-inventor of 3 granted US patents and 3 invention disclosures. He is the recipient of several competitive awards including the NSF CRII Award in 2017, the Best Graduate Research Assistant Award from the University of Maine in 2011, and the Best Paper Award from the IEEE/CANEUS Fly By Wireless Workshop in 2011. He is an IEEE Senior Member and served as TPC co-Chair, TPC Member, and Organizing Committee Member of several IEEE conferences including CISS 2015, FBW 2010, WiSEE 2014-19, CCNC2017-20, Microsoft CMT, VTC2017-19, PiMRC2017-18, SECON2018, and WiOPT 2016.
Title:“Multiple Instances Learning (MIL): A Weekly Supervised Approach to High-Resolution Medical Imaging”
Abstract:At the heart of many diagnostic systems lies biomedical imaging, which often involves processing large-scale, high-resolution images at megabyte and gigabyte scales. In today's landscape, most of these systems leverage deep learning (DL) architectures for tasks such as segmentation, object detection, and anomaly detection. However, a significant limitation of modern DL architectures is their restricted receptive field size, which limits their direct applicability to medical image processing. A common workaround is to divide these large images into smaller tiles and process them sequentially, a concept known as Whole Slide Imaging (WSI). This approach introduces a new challenge: the scarcity of human annotations for each image patch, as labels are generally provided at the image level, noting that labeling every instance can be impractical, expensive, and labor-intensive. A formal way to address this challenge is by utilizing Multiple Instance Learning (MIL), a weakly supervised method that offers a flexible framework for processing large, complex, and partially labeled datasets. In this context, each image tile is treated as a sample, while the entire high-resolution image is regarded as a bag of samples, with a label. This talk will delve into recent advancements in addressing these challenges, with a focus on improving weakly supervised image processing by enhancing the contribution of highly diverse instances. We will conclude by presenting our newly developed method, DGR-MIL, which aims to improve MIL performance by accounting for the inherent diversity among instances, while also distinguishing between positive and negative samples. Experimental results on two WSI benchmarks demonstrate that the proposed framework outperforms competing MIL aggregation methods. Please view our BDSI Seminar Calendar https://www.cs.clemson.edu/bdsi/calendar.html for upcoming seminars, journal club presentations and more ways to access the seminar.