W14. Deep Learning for Image Analysis
TUESDAY, october 6 | 12:30 - 3:30 PM
ABOUT THIS WORKSHOP:
Many of the recent success stories in Artificial Intelligence (AI) have leveraged a class of machine learning methods called deep learning. The initial wave of applications in this area has been on image understanding focused on the ability to recognize objects in images. The deep learning methods are replacing decades of image processing and computer vision methods that rely on models or feature engineering with a data-driven approach. The increase in accuracy brought by deep learning over the last few years has enabled their application to self-driving cars, manufacturing, astronomy, and a range of biomedical imaging problems. Participants of this introductory workshop will learn conceptually about how deep learning and specifically how convolutional neural networks (CNNs) work. We will discuss the appropriate space of problems for them within the pharmaceutical and biotech fields.
TOPICS TO BE COVERED:
- Differences between deep learning and traditional image processing pipelines
- Key advances in technology and methods for deep learning and CNNs
- Primary deep learning architectures and associated techniques
- Object recognition within images
- Transfer learning
- Application areas for CNNs in the industry
12:30 pm Workshop Begins
1:45 Refreshment Break
3:30 Workshop Ends
Peter Henstock, PhD, AI & Machine Learning Lead, Pfizer
Peter Henstock is working to transform Pfizer using AI and Machine Learning. He is the Machine Learning and Technical Lead in Pfizer’s Digital group based in Massachusetts. He holds a PhD in Artificial Intelligence from Purdue University and Master’s degrees in Biology, Software Engineering, Statistics, Applied Linguistics, and Image Processing. Before joining Pfizer, Peter worked at MIT Lincoln Laboratory in image processing and computational linguistics. He also teaches graduate level Machine Learning & Data Mining and Software Engineering at Harvard University.
Chao-Hui Huang, PhD, Senior Principal Scientist, Quantitative Image Analysis Computational Biology, Oncology Research Unit, Pfizer
Dr. Huang is currently a senior principal scientist in Pfizer. Previously he was with Merck/MSD Research Lab (MRL), Merck/MSD & Co. and Agency for Science, Technology and Research (A*STAR), Singapore. He has more than 20-year-rcperience in the fields of computer vision, machine learning and their applications of microscopic image, biomedical image and big unconstructed data analysis, such as pathological image analysis, immunohistochemistry data analysis, in-vivo image analysis and bioinformatics. He has been awarded 2 US/PCT patents. Until 2020, he has published more than 32 papers/book chapters in international journals and conferences.
Back to Workshops