© Fraunhofer MEVIS Meeting of the Working Group VCBM, 3. September 2013, Vienna Frank Heckel Fraunhofer MEVIS, Bremen, Germany | Innovation Center Computer.

Slides:



Advertisements
Similar presentations
Institute for vision and graphics university of siegen, germany High-Level User Interfaces for Transfer Function Design with Semantics Christof Rezk Salama.
Advertisements

Tooth Segmentation on Dental Meshes Using Morphologic Skeleton
1 Design, Prototyping, and Evaluation in Developing Countries Jen Mankoff, Assistant Professor EECS.
A System to Generate Test Data and Symbolically Execute Programs Lori A. Clarke September 1976.
SOFTWARE TESTING. INTRODUCTION  Software Testing is the process of executing a program or system with the intent of finding errors.  It involves any.
COMPUTER-AIDED SURGICAL PLANNING AND PROCEDURES A.Schaeffer; PolyDimensions GmbH, Bickenbach.
Test Automation for Verifying Software’s Detectability for Rule Violations Name: Zhishuai Yao Supervisor: Pro. Jukka Manner Place: Varian Medical Systems.
Data Mining Methodology 1. Why have a Methodology  Don’t want to learn things that aren’t true May not represent any underlying reality ○ Spurious correlation.
© Fraunhofer MEVIS Toward Automated Validation of Sketch-based 3D Segmentation Editing Tools Frank Heckel 1, Momchil I. Ivanov 2, Jan H. Moltz 1, Horst.
MT Evaluation: Human Measures and Assessment Methods : Machine Translation Alon Lavie February 23, 2011.
Tailoring Needs Chapter 3. Contents This presentation covers the following: – Design considerations for tailored data-entry screens – Design considerations.
D contour based local manual correction of liver segmentations1Institute for Medical Image Computing 3D contour based local manual correction.
Overview Introduction Variational Interpolation
15 February Partial volume correction for liver metastases and lymph nodes1Institute for Medical Image Computing/16SPIE 2010 Partial volume correction.
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 7: Expert Systems and Artificial Intelligence Decision Support.
Evaluating the Quality of Image Synthesis and Analysis Techniques Matthew O. Ward Computer Science Department Worcester Polytechnic Institute.
Law Enforcement Resource Allocation (LERA) Visualization System Michael Welsman-Dinelle April Webster.
The Calibration Process
Principle of Functional Verification Chapter 1~3 Presenter : Fu-Ching Yang.
Interactive Optimization by Genetic Algorithms Cases: Lighting Patterns and Image Enhancement Janne Koljonen Electrical Engineering and Automation, University.
For Better Accuracy Eick: Ensemble Learning
SEM Metrology Today the SEM is used in many fields as a metrology tool. Measurements are made from: –the screen image –directly from the micrographs –video.
1 CS101 Introduction to Computing Lecture 24 Design Heuristics.
Data Mining Techniques
© Fraunhofer MEVIS , Heidelberg Collaboratory for Image Processing Frank Heckel, PhD Software Support for Oncological Therapy Response Assessment.
An Interactive Segmentation Approach Using Color Pre- processing Marisol Martinez Escobar Ph.D Candidate Major Professor: Eliot Winer Department of Mechanical.
Multigenerational Analysis And Visualization of Large 3D Vascular Images Shu-Yen Wan Department of Information Management, Chang Gung University, Taiwan,
Sketch-Based Interactive Segmentation and Segmentation Editing for Oncological Therapy Monitoring Frank Heckel March 17, 2015 BVM-Award 2015 – PhD Thesis.
Where Innovation Is Tradition SYST699 – Spec Innovations Innoslate™ System Engineering Management Software Tool Test & Analysis.
Slide - 1 Confidential. Slide - 2 Confidential Slide - 3 Confidential Interalgorithm Study using CT Images of synthetic nodules……….
Design Science Method By Temtim Assefa.
Estimating the tumor-breast volume ratio from mammograms Jorge Rodríguez*, Pedro Linares*, Eduardo Urra*, Daniella Laya*, Felipe Saldivia , Aldo Reigosa.
EE 492 ENGINEERING PROJECT LIP TRACKING Yusuf Ziya Işık & Ashat Turlibayev Yusuf Ziya Işık & Ashat Turlibayev Advisor: Prof. Dr. Bülent Sankur Advisor:
Reconstructing 3D mesh from video image sequences supervisor : Mgr. Martin Samuelčik by Martin Bujňák specifications Master thesis
Interactive surface reconstruction on triangle meshes with subdivision surfaces Matthias Bein Fraunhofer-Institut für Graphische Datenverarbeitung IGD.
Foundations of Software Testing Chapter 5: Test Selection, Minimization, and Prioritization for Regression Testing Last update: September 3, 2007 These.
Bug Localization with Machine Learning Techniques Wujie Zheng
Knowledge-Based Breast Cancer Prognosis Olvi Mangasarian UW Madison & UCSD La Jolla Edward Wild UW Madison Computation and Informatics in Biology and Medicine.
Registration of functional PET and structural MR images PVEOut satellite meeting Budapest, June 11 th 2004 Peter Willendrup & Claus Svarer Neurobiology.
AUTOMATIZATION OF COMPUTED TOMOGRAPHY PATHOLOGY DETECTION Semyon Medvedik Elena Kozakevich.
Radiometric Correction and Image Enhancement Modifying digital numbers.
Feature based deformable registration of neuroimages using interest point and feature selection Leonid Teverovskiy Center for Automated Learning and Discovery.
A Segmentation Algorithm Using Dyadic Wavelet Transform and the Discrete Dynamic Contour Bernard Chiu University of Waterloo.
MedIX – Summer 07 Lucia Dettori (room 745)
AdvisorStudent Dr. Jia Li Shaojun Liu Dept. of Computer Science and Engineering, Oakland University Automatic 3D Image Segmentation of Internal Lung Structures.
BOĞAZİÇİ UNIVERSITY DEPARTMENT OF MANAGEMENT INFORMATION SYSTEMS MATLAB AS A DATA MINING ENVIRONMENT.
Review: Neural Network Control of Robot Manipulators; Frank L. Lewis; 1996.
A New Method for Automatic Clothing Tagging Utilizing Image-Click-Ads Introduction Conclusion Can We Do Better to Reduce Workload?
Visualization of Tumors in 4D Medical CT Datasets Visualization of Tumors in 4D Medical CT Datasets Burak Erem 1, David Kaeli 1, Dana Brooks 1, George.
The PET/CT Working Group: CT Segmentation Challenge Informatics Issues Multi-site algorithm comparison Task: CT-based lung nodule segmentation.
CSC400W Honors Project Proposal Understanding ocean surface features from satellite images Jared Tilanus Nemanja Spasic.
Flair development for the MC TPS Wioletta Kozłowska CERN / Medical University of Vienna.
Kim HS Introduction considering that the amount of MRI data to analyze in present-day clinical trials is often on the order of hundreds or.
Shadow Detection in Remotely Sensed Images Based on Self-Adaptive Feature Selection Jiahang Liu, Tao Fang, and Deren Li IEEE TRANSACTIONS ON GEOSCIENCE.
Software Testing By Souvik Roy. What is Software Testing? Executing software in a simulated or real environment, using inputs selected somehow.
Exposure Prediction and Measurement Error in Air Pollution and Health Studies Lianne Sheppard Adam A. Szpiro, Sun-Young Kim University of Washington CMAS.
Establishing by the laboratory of the functional requirements for uncertainty of measurements of each examination procedure Ioannis Sitaras.
CIRP Annals - Manufacturing Technology 60 (2011) 1–4 Augmented assembly technologies based on 3D bare-hand interaction S.K. Ong (2)*, Z.B. Wang Mechanical.
Information Aids for Diagnosis Tasks Based on Operators’ Strategies 김 종 현.
CircuitBoard Shane Zamora Eyrún A. Eyjólfsdóttir University of California, Santa Barbara Department of Computer Science Sketch-Based Logic Circuit Design.
Introduction to Machine Learning, its potential usage in network area,
Introduction Complex Ear Anatomy Ossicles Ossicels = Gehörknöchelchen
Dynamic management of segmented structures in 3D Slicer
You Zhang, Jeffrey Meyer, Joubin Nasehi Tehrani, Jing Wang
Wrapping up prototyping
The Calibration Process
Chapter 6 Calibration and Application Process
Automatic Digitizing.
CNRS applications in medical imaging
Developing Tasks This slide deck was adapted by Caitlin Kelleher based on the original by Saul Greenberg. (Thank you Saul)
Presentation transcript:

© Fraunhofer MEVIS Meeting of the Working Group VCBM, 3. September 2013, Vienna Frank Heckel Fraunhofer MEVIS, Bremen, Germany | Innovation Center Computer Assisted Surgery, Leipzig, Germany Sketch-Based Segmentation Editing for Oncological Therapy Monitoring

© Fraunhofer MEVIS 1 / 22 Why do we need efficient segmentation editing tools? Segmentation is one of the essential tasks in medical image analysis Many sophisticated automatic segmentation algorithms exist … … which might fail in some cases (low contrast, noise, biological variability) What to do? Manual segmentation?  Takes too long Different algorithm?  Might fail as well Locally correct the error!

© Fraunhofer MEVIS 2 / 22 The Segmentation Editing Process

© Fraunhofer MEVIS 3 / 22 What makes segmentation editing a difficult problem? Requirements: Intuitive interaction in 2D – estimate the user’s intention in 3D Local modifications Real-time feedback Provide a general tool (for different objects and modalities) Be independent of the preceding automatic algorithm The user expects the tool to allow him or her to correct all errors With only a few steps! The segmentation problems are typically hard

© Fraunhofer MEVIS 4 / 22 Editing Approaches Parameter tuning Direct interaction Moderate performance requirement High reproducibility Incorporate user information Indirect interaction Medium performance requirement Medium reproducibility Low-level manual correction Direct interaction Real-time performance requirement Low reproducibility High-level manual correction Indirect interaction Real-time performance requirement Medium reproducibility Guide initial algorithmDedicated editing tools

© Fraunhofer MEVIS 5 / 22 Sketch-Based Editing in 2D User inputCorrection resultEdited region Part containing the center of gravity

© Fraunhofer MEVIS 6 / 22 Sketch-Based Editing in 2D Possible Interactions add remove add + remove replace

© Fraunhofer MEVIS 7 / 22 Sketch-Based Editing in 3D The Correction Depth Add/Remove:Replace:

© Fraunhofer MEVIS 8 / 22 Sketch-Based Editing in 3D Image-Based Extrapolation

© Fraunhofer MEVIS 9 / 22 Sketch-Based Editing in 3D Image-Based Extrapolation s

© Fraunhofer MEVIS 10 / 22 Sketch-Based Editing in 3D Image-Independent Extrapolation

© Fraunhofer MEVIS 11 / 22 Sketch-Based Editing in 3D Previously performed corrections should be part of the new surface Keep all user-inputs and use them for reconstruction Major issue: Contradictory inputs Ignore constraints from previous contours within a specific range: Image-Independent Extrapolation 1 st input2 nd input

© Fraunhofer MEVIS 12 / 22 Sketch-Based Editing in 3D Image-Independent Extrapolation

© Fraunhofer MEVIS 13 / 22 Qualitative Evaluation Editing is a dynamic, user- driven process User studies are the most important tool for evaluation Quality of editing tools is highly subjective Subjective quality suffers from bad intermediate results (Vague) criteria: Accuracy Efficiency Repeatability visually performed by the user intended result that the user only has in mind

© Fraunhofer MEVIS 14 / 22 Qualitative Evaluation Rating Scheme for Accuracy and Efficiency sufficient insufficient

© Fraunhofer MEVIS 15 / 22 Qualitative Evaluation Results Editing rating score: 131 representative tumor segmentations in CT (lung nodules, liver metastases, lymph nodes) 5 radiologists with different level of experience

© Fraunhofer MEVIS 16 / 22 Quantitative Evaluation

© Fraunhofer MEVIS 17 / 22 Quantitative Evaluation Results

© Fraunhofer MEVIS 18 / 22 Simulation-Based Evaluation Problem: High effort and bad reproducibility of user studies Idea: Replace user by a simulation Benefits: Objective and reproducible validation Objective comparison Improved regression testing Better parameter tuning

© Fraunhofer MEVIS 19 / 22 Simulation-Based Evaluation Simulation Step 1: Find most probably corrected 3D error Step 2: Select slice and view where the error is most probably corrected Step 3: Generate user-input for sketching Step 4: Apply editing algorithm

© Fraunhofer MEVIS 20 / 22 Simulation-Based Evaluation Results

© Fraunhofer MEVIS 21 / 22 Takeaway Message The segmentation problem is not solved yet Segmentation editing is an indispensable step in the segmentation process Particularly for clinical routine, editing is a mandatory feature and not only nice to have Efficient editing in 3D is challenging and not much work has been done in this field so far Sketching has shown to be an intuitive interface in 2D The presented methods provide efficient 3D editing tools for tumor segmentation

© Fraunhofer MEVIS 22 / 22 What's next? Editing for complex objects with irregular shapes Support additional user inputs Combine image-based and image-independent approaches Further improve the evaluation (measures, simulation) Find the most efficient human-computer interfaces for editing

© Fraunhofer MEVIS Thank you!