Direct image-analysis methods for surgical simulation and mixed - PowerPoint PPT Presentation

Direct image-analysis methods for surgical simulation and mixed meshfree methods Jack S. Hale Marie Curie Postdoctoral Fellow Research Unit in Engineering Science Université du Luxembourg Collaborators: Stephane P. A. Bordas, Universite du Luxembourg/Cardiff University Pierre Kerfriden, Cardiff University Juan J. Rodenas Garcia, Universidad Politecnica de Valencia � Pedro M. Baiz, Imperial College London Alejandro Ortiz-Bernardin, Universidad de Chile Christian J. Cyron, Yale University Université Pierre-et-Marie-Curie Institut Jean le Rond d’Alembert 26th May 1400

An Overview

The problems and my research Mixed Meshfree variational methods forms Numerical Direct image- locking in Mesh distortion in analysis transition incompressible hyperelastic elasticity, plates and for surgical problems shells, Cosserat simulators. elasticity Enriched implicit Patch projection techniques boundary methods

Project 1: Direct image to mesh analysis for surgical simulation

Enriched and implicit High Performance boundary methods Computing Surgical Simulation Model-order Real-time simulation reduction

RealTCut and SHACRA Inria

How can we move from an image… COLONIX, OSIRIX

…or perhaps a series of images… Source: COLONIX, OSIRIX

Pipelines to analysis Traditional Implicit Boundary } } Acquire images Acquire images Implicit Segment images Segment images { NURBS Mesh Surfaces Implicit Explicit Mesh Volume Perform analysis Perform analysis

Nested Octree Discretisation Geometry O g O d M

How to transfer geometric information back to the discretisation? Enrichment E [ V p g V p d � h d ( O d ) h g ( O g )] V p g V p d p d > p g h d ( O d ) h g ( O g ) h d > h g p g = 1

Octree Level 5/Level 3

Surface 6a2e86c

Features • 2D and 3D problems on the same code-path. • parallel hybrid MPI/TBB assembly and solution. • fast and robust computational geometry using CGAL. • automatic Delaunay tessellation of integration subdomains. • completely separate representation of discretisation and geometry via nested octree data structures. • constructs to represent soft and hard segmentations of image data. • implicit representation via level-sets, inside-outside functions. • independent hpe-type adaptivity on discretisation and geometry. • fast refinement and coarsening operations.

Outlook • We are developing a cartesian grid implicit boundary/enriched finite element method toolkit within deal.ii. • By uncoupling discretisation and geometry we hope to produce a method particularly suited to image-based analysis. • Many challenges ahead, particularly with imposition of Dirichlet boundary conditions on hyperelastic soft-tissue model.

Project 2: Mesh-based and Meshfree Partitions of Unity

Partition of Unity N � u h ( x ) = φ i u i i =1 N � φ i = 1 i =1

�� Finite elements 1 . 0 0 . 8 0 . 6 φ 0 . 4 0 . 2 0 . 0 0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0 x ��

�� Mesh-based Partition of Unity

�� Finite Element Method Reference Mesh

�� Meshfree-based Partition of Unity

Meshfree partition of unity is constructed in global space Good: Not so good: improved approximation expensive. • • properties. extra flexibility also means extra • high-order continuity. complexity: • less dofs for given accuracy. integration. � • • less sensitive to poor quality adaptivity/error measures. • • node distributions. � mathematical proof. • eases mesh generation. stability. � • • boundary conditions. � • �

Direct image-analysis methods for surgical simulation and mixed - PowerPoint PPT Presentation

Direct image-analysis methods for surgical simulation and mixed meshfree methods Jack S. Hale Marie Curie Postdoctoral Fellow Research Unit in Engineering Science Universit du Luxembourg Collaborators: Stephane P. A. Bordas, Universite du

Direct Cardiac Image Analysis without Segmentation Dr. Shuo Li the Digital Imaging Group (DIG)

SURGICAL SKILLS EDUCATION IN UROLOGY DOES IT MATTER? Conflict of Interest Statement No

Perfect simulation for image analysis Mark Huber Fletcher Jones Foundation Associate Professor of

Surgical Simulation: Surgical Simulation: We dont need simulation. We dont need

Time Series Based Semi-Analytical Solution of Power Systems and its Application in Direct Methods

Why Bayesian methods in Simulation? Simulation Simulation Model Inputs BAYESIAN IDEAS

DESIGN & ANALYSIS METHODS DESIGN & ANALYSIS METHODS FOR DESIGN & ANALYSIS METHODS

Testing/Simulation Formal Analysis Real System Formal Model Partial coverage Complete coverage

Image Diagnostics, Inc. Manufacturer of quality products for surgical, interventional and

Direct Methods for Solving Linear Systems Linear Systems of Equations Numerical Analysis (9th

Direct Methods for Solving Linear Systems Pivoting Strategies Numerical Analysis (9th Edition) R

Direct Methods for Solving Linear Systems Matrix Factorization Numerical Analysis (9th Edition)

Where are we? Informatics 2D Reasoning and Agents Semester 2, 20192020 Last time . . .

Edge Detection (points and lines too) Image analysis Image Analysis Derive

CMOS Image Sensor Simulation 2D and 3D Simulation Basic Structure and operation of CMOS Image

Chapter 6 Direct Methods for Solving Linear Systems Per-Olof Persson persson@berkeley.edu

ray tracing 1 image formation 2 image formation 3 rendering computational simulation of

Reduction Methods for Multiband Image Analysis F. Flitti 1 , C. Collet 1 and F.Bonnarel 2 1 LSIIT,

Simulation, Measurement and Analysis of Photoemission Simulation, Measurement and Analysis of

Direct Search Methods (nongradient methods) 1. Random search methods 2. Univariate method (one

Haptics for Surgical Simulation CPSC 599.86 / 601.86 Sonny Chan University of Calgary Motivation

Analyzing Simulated Data Matthew Turk There is only one sky. (but there are many simulation

Graph-based Methods in Pattern Recognition and Document Image Analysis (G M PR D I A ) Tutorial

Haptics for Surgical Simulation CPSC 599.86 / 601.86 Sonny Chan University of Calgary A few

Direct image-analysis methods for surgical simulation and mixed - PowerPoint PPT Presentation

Direct image-analysis methods for surgical simulation and mixed meshfree methods Jack S. Hale Marie Curie Postdoctoral Fellow Research Unit in Engineering Science Universit du Luxembourg Collaborators: Stephane P. A. Bordas, Universite du

Direct Cardiac Image Analysis without Segmentation Dr. Shuo Li the Digital Imaging Group (DIG)

SURGICAL SKILLS EDUCATION IN UROLOGY DOES IT MATTER? Conflict of Interest Statement No

Perfect simulation for image analysis Mark Huber Fletcher Jones Foundation Associate Professor of

Surgical Simulation: Surgical Simulation: We dont need simulation. We dont need

Time Series Based Semi-Analytical Solution of Power Systems and its Application in Direct Methods

Why Bayesian methods in Simulation? Simulation Simulation Model Inputs BAYESIAN IDEAS

DESIGN &amp; ANALYSIS METHODS DESIGN &amp; ANALYSIS METHODS FOR DESIGN &amp; ANALYSIS METHODS

Testing/Simulation Formal Analysis Real System Formal Model Partial coverage Complete coverage

Image Diagnostics, Inc. Manufacturer of quality products for surgical, interventional and

Direct Methods for Solving Linear Systems Linear Systems of Equations Numerical Analysis (9th

Direct Methods for Solving Linear Systems Pivoting Strategies Numerical Analysis (9th Edition) R

Direct Methods for Solving Linear Systems Matrix Factorization Numerical Analysis (9th Edition)

Where are we? Informatics 2D Reasoning and Agents Semester 2, 20192020 Last time . . .

Edge Detection (points and lines too) Image analysis Image Analysis Derive

CMOS Image Sensor Simulation 2D and 3D Simulation Basic Structure and operation of CMOS Image

Chapter 6 Direct Methods for Solving Linear Systems Per-Olof Persson persson@berkeley.edu

ray tracing 1 image formation 2 image formation 3 rendering computational simulation of

Reduction Methods for Multiband Image Analysis F. Flitti 1 , C. Collet 1 and F.Bonnarel 2 1 LSIIT,

Simulation, Measurement and Analysis of Photoemission Simulation, Measurement and Analysis of

Direct Search Methods (nongradient methods) 1. Random search methods 2. Univariate method (one

Haptics for Surgical Simulation CPSC 599.86 / 601.86 Sonny Chan University of Calgary Motivation

Analyzing Simulated Data Matthew Turk There is only one sky. (but there are many simulation

Graph-based Methods in Pattern Recognition and Document Image Analysis (G M PR D I A ) Tutorial

Haptics for Surgical Simulation CPSC 599.86 / 601.86 Sonny Chan University of Calgary A few

DESIGN & ANALYSIS METHODS DESIGN & ANALYSIS METHODS FOR DESIGN & ANALYSIS METHODS