• Core tasks of bioimage analysis: detection, segmentation and tracking
• Capturing contextual interactions and prior knowledge
• End-to-end training and structured prediction

• Mathematical approaches to modeling shape
• Biologically meaningful correspondence
• Connecting form and function
  • Pattern formation in Drosophila development
  • Ear morphology and echolocation in bats
  [Ref 1], [Ref 2], [Ref 3], [Ref 4], [Ref 5], [Ref 6], [Ref 7], [Ref 8], [Ref 9], [Ref 10], [Ref 11].

One direction in medical image analysis is to effectively represent knowledge and efficiently extract biomedical information (such as a deformable shape) from medical images. In particular, machine learning techniques (supervised, weakly-supervised, and unsupervised) have played increasingly important role. The large-scale data learning and analysis have also recently played a significant role in medical imaging.

The goal of this lecture is to provide a comprehensive assessment of discriminative learning techniques used for medical imaging applications such as anatomical structure detection and segmentation, image categorization, etc. Learning from an annotated dataset the covers the uncertainties involved in the applications, these techniques are able to derive compact descriptions between the image and knowledge and gain improvements in performance and speed when compared with conventional algorithms without using learning.

Coverage: Applications of supervised and semi-supervised learning in recent medical imaging applications

• A brief introduction to Pattern Theory and practice
• Demonstrations of Typical patterns in digitized life science data sets: ubiquitous, low-rank, deformed regular patterns
• Symmetry group-based regularity space - a novel, computable model
• Computational challenges: why is it hard for computer vision algorithms to discover real world, free-form symmetries?

• On Growth and Form, D'Arcy Thompson, Cambridge University Press, Jul 31, 1992 - 368 pages
• Pattern Theory: From Representation to Inference, Ulf Grenander and Michael Miller, Oxford University Press, USA (February 8, 2007)
• Computational Symmetry in Computer Vision and Computer Graphics. Yanxi Liu, Hagit Hel-Or, Craig S. Kaplan, and Luc Van Gool. Foundations and Trends in Computer Graphics and Vision 2010. Volume 5, Number 1-2, Pages 199. [Project Page]
• Computational Symmetry, Yanxi Liu, Symmetry 2000, Portland Press, London, Vol. 80/1, January, 2002, pp. 231 - 245.
• Symmetry Detection from Real World Images - A Competition, US NSF-Funded CVPR 2011 Workshop

• Discriminative feature subset selection (off-line, on-line) [13, 14, 19]
• Curved glide reflection symmetry detection [1, 2]
• Skewed rotation symmetry detection [3]
• Texture regularity discovery (translation symmetry) [4-6]

• 2D/3D human identification, expression/gender classification [7, 8]
• Quantified patterns (firing fields of grid cells)[9, 10]
• Tracking patterns (gated cardiac MRI)[11-12]
• Evaluation of Scoliosis[1]
• Computer aided diagnosis for neurodegenerative diseases (Alzheimer's Disease, Schizophrenia...)[13, 14]
• Automatic Detection of Midsagittal Plane from Volumetric Neuroimages[20]
• Brain tumor detection and segmentation[15]
• Zebra fish (wild versus mutant)[16, 17]
• Human gaits/dance[6, 18]

1. Curved Glide-Reflection Symmetry Detection, PDF, Project Page
2. Curved Reflection Symmetry Detection with Self-validation, PDF, Project Page
3. Skewed Rotation Symmetry Group Detection, PDF, Project Page
4. Deformed Lattice Detection in Real-World Images using Mean-Shift Belief Propagation, PDF, Project Page
5. Discovering Texture Regularity as a Higher-Order Correspondence Problem, PDF
6. A Computational Model for Periodic Pattern Perception Based on Frieze and Wallpaper Groups, PDF
7. Facial Asymmetry Quantification for Expression Invariant Human Identification, PDF
8. A Quantified Study of Facial Asymmetry in 3D Faces, PDF
9. Near-Regular Texture Analysis and Manipulation, PDF
10. Quantified Symmetry for Entorhinal Spatial Maps, PDF
11. A Lattice-based MRF Model for Dynamic Near-regular Texture Tracking, PDF
12. Multi-Target Tracking of Time-Varying Spatial Patterns, PDF, Project Page, Video
13. Discovery of "Biomarkers" for Alzheimer's Disease Prediction from Structural MR Images, PDF, Project Page
14. Discriminative MR Image Feature Analysis for Automatic Schizophrenia and Alzheimer's Disease Classification, PDF
15. Statistical Asymmetry-based Brain Tumor Segmentation from 3D MR Images (2012), PDF, Project Page
16. Towards Efficient Automated Characterization of Irregular Histology Images via Transformation to Frieze-Like Patterns, PDF
17. Automatic Lattice Detection in Near-Regular Histology Array Images, PDF
18. Gait Sequence Analysis using Frieze Patterns (2002), PDF
19. On-Line Selection of Discriminative Tracking Features, PDF
20. Robust Midsagittal Plane Extraction from Normal and Pathological 3D Neuroradiology Images, PDF

Now that you have teams, begin brainstorming on a potential collaboration. Explore ideas and themes from the workshop and from your home research lab. Start to put together the pitch for tomorrow!