RegistrationShop - An Interactive 3D Medical Volume Registration System

Noeska Smit
Noeska SmitPhD student at Delft University of Technology
RegistrationShop: 
An Interactive 3D Medical Volume Registration System 
Noeska N. Smit1, Berend Klein Haneveld1, 
Marius Staring2, Elmar Eisemann1, 
Charl P. Botha3 and Anna Vilanova1 
1 Computer Graphics and Visualization group, Intelligent Systems, TU Delft, The Netherlands 
2 Division of Image Processing, Department of Radiology, LUMC, The Netherlands 
3 vxlabs, South Africa 
RegistrationShop Noeska Smit
Pre- and post treatment 
RegistrationShop Noeska Smit
Multiple modalities 
RegistrationShop Noeska Smit 
http://www.osirix-viewer.com/Snapshots.html
Atlas-based Segmentation 
RegistrationShop Noeska Smit
Registration 
RegistrationShop Noeska Smit
Registration Functionality 
• AMIDE 
• AMILab 
• FusionSync 
• Ezys 
• Elastix 
• MITK 
• Mirada Medical 
• PMOD 
• Slicer 
• VolView 
• Voreen 
• … 
RegistrationShop Noeska Smit
Slicer 
RegistrationShop Noeska Smit
Elastix 
RegistrationShop Noeska Smit
Comparison 
• Typically only 2D visualizations of the registration result or 
missing required transformation tools 
• Not always suitable for non-image processing experts 
RegistrationShop Noeska Smit
What if… 
• simple interaction techniques 
• + real-time 3D visual feedback 
• = simplified registration process for novice 
users? 
RegistrationShop Noeska Smit
Landmark placement 
Corresponding landmark pairs: 
– Method 1: surface picker 
RegistrationShop Noeska Smit
Landmark placement 
Corresponding landmark pairs: 
– Method 1: surface picker 
– Method 2: two-step picker 
RegistrationShop Noeska Smit
Simple visualization techniques 
Direct volume rendering: 
Double thresholds MIP 
RegistrationShop Noeska Smit
Multi-volume rendering 
Real-time visual feedback on the current registration 
result: 
THR/THR MIP/THR MIP/MIP 
RegistrationShop Noeska Smit
RegistrationShop 
RegistrationShop Noeska Smit
Implementation 
• Source code freely available via GitHub: 
• github.com/berendkleinhaneveld/Registrationshop 
• Installable builds coming soon! 
RegistrationShop Noeska Smit
RegistrationShop 
RegistrationShop Noeska Smit
Evaluation 
• Four domain experts: 
– Medical radiation therapist 
– Radiologist 
– Medical imaging researcher: + registration frequency 
– Medical imaging researcher: - registration frequency 
• Two use-cases: 
– CT lung 
– Full body MRI 
RegistrationShop Noeska Smit
Evaluation Setup 
1. Prior experience interview 
2. Live demo: artificial dataset 
3. Performing registration of 
clinical datasets 
4. Survey 
RegistrationShop Noeska Smit
Evaluation Results 
• 3D visualizations 
• Transformation tools 
• Conclusion: 
– RegistrationShop: a promising volume registration tool 
with the potential to make registration more accessible to 
clinicians and researchers in the future. 
RegistrationShop Noeska Smit
Future Work 
• Visualizing registration quality 
• Visualizing non-rigid registration deformation field 
• Experimenting with compositing operations and 
transfer function definitions 
• Examining clinical potential in larger follow-up user 
study 
RegistrationShop Noeska Smit
Conclusion 
• RegistrationShop: medical volume registration 
system 
• Simple transformation techniques: 
– manual and landmark transformations 
• Instant visual feedback: 
– 2D/3D inspection of current registration result 
RegistrationShop Noeska Smit
Acknowledgements 
• This work is supported by the 
Technology Foundation STW 
via project 10903: 
“High-definition Atlas-based 
surgical planning for Pelvic Surgery” 
• CT lung datasets shown in this presentation are 
available from the Elastix test suite 
• We would like to thank our collaborators and 
domain experts for their participation in the user 
study 
RegistrationShop Noeska Smit
Thank you! 
@_Noeska_ , @berendkleinhane 
@medvisorg github.com/berendkleinhaneveld/Registrationshop 
RegistrationShop Noeska Smit
1 of 24

Recommended

What Is Visualization? by
What Is Visualization?What Is Visualization?
What Is Visualization?OneSpring LLC
807 views40 slides
Theius: A Streaming Visualization Suite for Hadoop Clusters by
Theius: A Streaming Visualization Suite for Hadoop ClustersTheius: A Streaming Visualization Suite for Hadoop Clusters
Theius: A Streaming Visualization Suite for Hadoop Clustersjtedesco5
926 views26 slides
Real-Time Analytics and Visualization of Streaming Big Data with JReport & Sc... by
Real-Time Analytics and Visualization of Streaming Big Data with JReport & Sc...Real-Time Analytics and Visualization of Streaming Big Data with JReport & Sc...
Real-Time Analytics and Visualization of Streaming Big Data with JReport & Sc...Mia Yuan Cao
1.4K views20 slides
Info vis 4-22-2013-dc-vis-meetup-shneiderman by
Info vis 4-22-2013-dc-vis-meetup-shneidermanInfo vis 4-22-2013-dc-vis-meetup-shneiderman
Info vis 4-22-2013-dc-vis-meetup-shneidermanUniversity of Maryland
1.2K views97 slides
Building a Big Data Pipeline by
Building a Big Data PipelineBuilding a Big Data Pipeline
Building a Big Data PipelineJesus Rodriguez
6.9K views52 slides
Information Visualization for Medical Informatics by
Information Visualization for Medical Informatics Information Visualization for Medical Informatics
Information Visualization for Medical Informatics University of Maryland
813 views51 slides

More Related Content

Viewers also liked

Towards Utilizing GPUs in Information Visualization by
Towards Utilizing GPUs in Information VisualizationTowards Utilizing GPUs in Information Visualization
Towards Utilizing GPUs in Information VisualizationNiklas Elmqvist
737 views22 slides
Text and text stream mining tutorial by
Text and text stream mining tutorialText and text stream mining tutorial
Text and text stream mining tutorialmgrcar
8.1K views88 slides
Presentation Brucon - Anubisnetworks and PTCoresec by
Presentation Brucon - Anubisnetworks and PTCoresecPresentation Brucon - Anubisnetworks and PTCoresec
Presentation Brucon - Anubisnetworks and PTCoresecTiago Henriques
3.9K views71 slides
In Memory Analytics with Apache Spark by
In Memory Analytics with Apache SparkIn Memory Analytics with Apache Spark
In Memory Analytics with Apache SparkVenkata Naga Ravi
3.5K views28 slides
Web 2 0 Projects Elementary by
Web 2 0 Projects ElementaryWeb 2 0 Projects Elementary
Web 2 0 Projects ElementaryCinci0987
295 views21 slides
An Introduction to Evaluation in Medical Visualization by
An Introduction to Evaluation in Medical VisualizationAn Introduction to Evaluation in Medical Visualization
An Introduction to Evaluation in Medical VisualizationNoeska Smit
483 views26 slides

Viewers also liked(7)

Towards Utilizing GPUs in Information Visualization by Niklas Elmqvist
Towards Utilizing GPUs in Information VisualizationTowards Utilizing GPUs in Information Visualization
Towards Utilizing GPUs in Information Visualization
Niklas Elmqvist737 views
Text and text stream mining tutorial by mgrcar
Text and text stream mining tutorialText and text stream mining tutorial
Text and text stream mining tutorial
mgrcar8.1K views
Presentation Brucon - Anubisnetworks and PTCoresec by Tiago Henriques
Presentation Brucon - Anubisnetworks and PTCoresecPresentation Brucon - Anubisnetworks and PTCoresec
Presentation Brucon - Anubisnetworks and PTCoresec
Tiago Henriques3.9K views
In Memory Analytics with Apache Spark by Venkata Naga Ravi
In Memory Analytics with Apache SparkIn Memory Analytics with Apache Spark
In Memory Analytics with Apache Spark
Venkata Naga Ravi3.5K views
Web 2 0 Projects Elementary by Cinci0987
Web 2 0 Projects ElementaryWeb 2 0 Projects Elementary
Web 2 0 Projects Elementary
Cinci0987295 views
An Introduction to Evaluation in Medical Visualization by Noeska Smit
An Introduction to Evaluation in Medical VisualizationAn Introduction to Evaluation in Medical Visualization
An Introduction to Evaluation in Medical Visualization
Noeska Smit483 views
Stream Processing with Kafka in Uber, Danny Yuan by confluent
Stream Processing with Kafka in Uber, Danny Yuan Stream Processing with Kafka in Uber, Danny Yuan
Stream Processing with Kafka in Uber, Danny Yuan
confluent7.5K views

Recently uploaded

Fleet Management Software in India by
Fleet Management Software in India Fleet Management Software in India
Fleet Management Software in India Fleetable
11 views1 slide
HarshithAkkapelli_Presentation.pdf by
HarshithAkkapelli_Presentation.pdfHarshithAkkapelli_Presentation.pdf
HarshithAkkapelli_Presentation.pdfharshithakkapelli
11 views16 slides
Consulting for Data Monetization Maximizing the Profit Potential of Your Data... by
Consulting for Data Monetization Maximizing the Profit Potential of Your Data...Consulting for Data Monetization Maximizing the Profit Potential of Your Data...
Consulting for Data Monetization Maximizing the Profit Potential of Your Data...Flexsin
15 views10 slides
SAP FOR TYRE INDUSTRY.pdf by
SAP FOR TYRE INDUSTRY.pdfSAP FOR TYRE INDUSTRY.pdf
SAP FOR TYRE INDUSTRY.pdfVirendra Rai, PMP
23 views3 slides
Dev-Cloud Conference 2023 - Continuous Deployment Showdown: Traditionelles CI... by
Dev-Cloud Conference 2023 - Continuous Deployment Showdown: Traditionelles CI...Dev-Cloud Conference 2023 - Continuous Deployment Showdown: Traditionelles CI...
Dev-Cloud Conference 2023 - Continuous Deployment Showdown: Traditionelles CI...Marc Müller
36 views83 slides
DSD-INT 2023 Thermobaricity in 3D DCSM-FM - taking pressure into account in t... by
DSD-INT 2023 Thermobaricity in 3D DCSM-FM - taking pressure into account in t...DSD-INT 2023 Thermobaricity in 3D DCSM-FM - taking pressure into account in t...
DSD-INT 2023 Thermobaricity in 3D DCSM-FM - taking pressure into account in t...Deltares
9 views26 slides

Recently uploaded(20)

Fleet Management Software in India by Fleetable
Fleet Management Software in India Fleet Management Software in India
Fleet Management Software in India
Fleetable11 views
Consulting for Data Monetization Maximizing the Profit Potential of Your Data... by Flexsin
Consulting for Data Monetization Maximizing the Profit Potential of Your Data...Consulting for Data Monetization Maximizing the Profit Potential of Your Data...
Consulting for Data Monetization Maximizing the Profit Potential of Your Data...
Flexsin 15 views
Dev-Cloud Conference 2023 - Continuous Deployment Showdown: Traditionelles CI... by Marc Müller
Dev-Cloud Conference 2023 - Continuous Deployment Showdown: Traditionelles CI...Dev-Cloud Conference 2023 - Continuous Deployment Showdown: Traditionelles CI...
Dev-Cloud Conference 2023 - Continuous Deployment Showdown: Traditionelles CI...
Marc Müller36 views
DSD-INT 2023 Thermobaricity in 3D DCSM-FM - taking pressure into account in t... by Deltares
DSD-INT 2023 Thermobaricity in 3D DCSM-FM - taking pressure into account in t...DSD-INT 2023 Thermobaricity in 3D DCSM-FM - taking pressure into account in t...
DSD-INT 2023 Thermobaricity in 3D DCSM-FM - taking pressure into account in t...
Deltares9 views
360 graden fabriek by info33492
360 graden fabriek360 graden fabriek
360 graden fabriek
info3349224 views
Unmasking the Dark Art of Vectored Exception Handling: Bypassing XDR and EDR ... by Donato Onofri
Unmasking the Dark Art of Vectored Exception Handling: Bypassing XDR and EDR ...Unmasking the Dark Art of Vectored Exception Handling: Bypassing XDR and EDR ...
Unmasking the Dark Art of Vectored Exception Handling: Bypassing XDR and EDR ...
Donato Onofri711 views
El Arte de lo Possible by Neo4j
El Arte de lo PossibleEl Arte de lo Possible
El Arte de lo Possible
Neo4j38 views
DSD-INT 2023 FloodAdapt - A decision-support tool for compound flood risk mit... by Deltares
DSD-INT 2023 FloodAdapt - A decision-support tool for compound flood risk mit...DSD-INT 2023 FloodAdapt - A decision-support tool for compound flood risk mit...
DSD-INT 2023 FloodAdapt - A decision-support tool for compound flood risk mit...
Deltares13 views
Elevate your SAP landscape's efficiency and performance with HCL Workload Aut... by HCLSoftware
Elevate your SAP landscape's efficiency and performance with HCL Workload Aut...Elevate your SAP landscape's efficiency and performance with HCL Workload Aut...
Elevate your SAP landscape's efficiency and performance with HCL Workload Aut...
HCLSoftware6 views
SUGCON ANZ Presentation V2.1 Final.pptx by Jack Spektor
SUGCON ANZ Presentation V2.1 Final.pptxSUGCON ANZ Presentation V2.1 Final.pptx
SUGCON ANZ Presentation V2.1 Final.pptx
Jack Spektor22 views
Software testing company in India.pptx by SakshiPatel82
Software testing company in India.pptxSoftware testing company in India.pptx
Software testing company in India.pptx
SakshiPatel827 views
DSD-INT 2023 3D hydrodynamic modelling of microplastic transport in lakes - J... by Deltares
DSD-INT 2023 3D hydrodynamic modelling of microplastic transport in lakes - J...DSD-INT 2023 3D hydrodynamic modelling of microplastic transport in lakes - J...
DSD-INT 2023 3D hydrodynamic modelling of microplastic transport in lakes - J...
Deltares9 views
Cycleops - Automate deployments on top of bare metal.pptx by Thanassis Parathyras
Cycleops - Automate deployments on top of bare metal.pptxCycleops - Automate deployments on top of bare metal.pptx
Cycleops - Automate deployments on top of bare metal.pptx
Generic or specific? Making sensible software design decisions by Bert Jan Schrijver
Generic or specific? Making sensible software design decisionsGeneric or specific? Making sensible software design decisions
Generic or specific? Making sensible software design decisions
DSD-INT 2023 European Digital Twin Ocean and Delft3D FM - Dols by Deltares
DSD-INT 2023 European Digital Twin Ocean and Delft3D FM - DolsDSD-INT 2023 European Digital Twin Ocean and Delft3D FM - Dols
DSD-INT 2023 European Digital Twin Ocean and Delft3D FM - Dols
Deltares7 views
DSD-INT 2023 - Delft3D User Days - Welcome - Day 3 - Afternoon by Deltares
DSD-INT 2023 - Delft3D User Days - Welcome - Day 3 - AfternoonDSD-INT 2023 - Delft3D User Days - Welcome - Day 3 - Afternoon
DSD-INT 2023 - Delft3D User Days - Welcome - Day 3 - Afternoon
Deltares13 views
DSD-INT 2023 Leveraging the results of a 3D hydrodynamic model to improve the... by Deltares
DSD-INT 2023 Leveraging the results of a 3D hydrodynamic model to improve the...DSD-INT 2023 Leveraging the results of a 3D hydrodynamic model to improve the...
DSD-INT 2023 Leveraging the results of a 3D hydrodynamic model to improve the...
Deltares6 views

RegistrationShop - An Interactive 3D Medical Volume Registration System

  • 1. RegistrationShop: An Interactive 3D Medical Volume Registration System Noeska N. Smit1, Berend Klein Haneveld1, Marius Staring2, Elmar Eisemann1, Charl P. Botha3 and Anna Vilanova1 1 Computer Graphics and Visualization group, Intelligent Systems, TU Delft, The Netherlands 2 Division of Image Processing, Department of Radiology, LUMC, The Netherlands 3 vxlabs, South Africa RegistrationShop Noeska Smit
  • 2. Pre- and post treatment RegistrationShop Noeska Smit
  • 3. Multiple modalities RegistrationShop Noeska Smit http://www.osirix-viewer.com/Snapshots.html
  • 6. Registration Functionality • AMIDE • AMILab • FusionSync • Ezys • Elastix • MITK • Mirada Medical • PMOD • Slicer • VolView • Voreen • … RegistrationShop Noeska Smit
  • 9. Comparison • Typically only 2D visualizations of the registration result or missing required transformation tools • Not always suitable for non-image processing experts RegistrationShop Noeska Smit
  • 10. What if… • simple interaction techniques • + real-time 3D visual feedback • = simplified registration process for novice users? RegistrationShop Noeska Smit
  • 11. Landmark placement Corresponding landmark pairs: – Method 1: surface picker RegistrationShop Noeska Smit
  • 12. Landmark placement Corresponding landmark pairs: – Method 1: surface picker – Method 2: two-step picker RegistrationShop Noeska Smit
  • 13. Simple visualization techniques Direct volume rendering: Double thresholds MIP RegistrationShop Noeska Smit
  • 14. Multi-volume rendering Real-time visual feedback on the current registration result: THR/THR MIP/THR MIP/MIP RegistrationShop Noeska Smit
  • 16. Implementation • Source code freely available via GitHub: • github.com/berendkleinhaneveld/Registrationshop • Installable builds coming soon! RegistrationShop Noeska Smit
  • 18. Evaluation • Four domain experts: – Medical radiation therapist – Radiologist – Medical imaging researcher: + registration frequency – Medical imaging researcher: - registration frequency • Two use-cases: – CT lung – Full body MRI RegistrationShop Noeska Smit
  • 19. Evaluation Setup 1. Prior experience interview 2. Live demo: artificial dataset 3. Performing registration of clinical datasets 4. Survey RegistrationShop Noeska Smit
  • 20. Evaluation Results • 3D visualizations • Transformation tools • Conclusion: – RegistrationShop: a promising volume registration tool with the potential to make registration more accessible to clinicians and researchers in the future. RegistrationShop Noeska Smit
  • 21. Future Work • Visualizing registration quality • Visualizing non-rigid registration deformation field • Experimenting with compositing operations and transfer function definitions • Examining clinical potential in larger follow-up user study RegistrationShop Noeska Smit
  • 22. Conclusion • RegistrationShop: medical volume registration system • Simple transformation techniques: – manual and landmark transformations • Instant visual feedback: – 2D/3D inspection of current registration result RegistrationShop Noeska Smit
  • 23. Acknowledgements • This work is supported by the Technology Foundation STW via project 10903: “High-definition Atlas-based surgical planning for Pelvic Surgery” • CT lung datasets shown in this presentation are available from the Elastix test suite • We would like to thank our collaborators and domain experts for their participation in the user study RegistrationShop Noeska Smit
  • 24. Thank you! @_Noeska_ , @berendkleinhane @medvisorg github.com/berendkleinhaneveld/Registrationshop RegistrationShop Noeska Smit

Editor's Notes

  1. Imagine you have a pre- and post-treatment scan of a patient and we want to examine the treatment effects. There is one small problem, the patient is not in the exact same position in both scans Wouldn’t it be great if we can just register these two scans to spot the differences?
  2. The same holds for combining two scans from different modalities to enrich them with additional information? T1: anatomy structural information DWI: water diffusion which can indicate tumor characteristics
  3. What about using an atlas to segment an unseen dataset?
  4. We need an easy way to register all these dataset pairs together
  5. Many tools that perform registrations exist, but they are often complex for novice users
  6. We could try to manually align the datasets using rigid transformations in a tool like Slicer But is using sliders and 2D overlays really the easiest way to do this? Or to place corresponding landmark pairs and align the scans based on these But we only see the result of the transformation once all landmark pairs are set
  7. We could use a command-line tool such as Elastix to perform deformable registrations But it has no direct visual feedback on intermediate registration results and the learning curve for novice users is steep
  8. When we compare these tools, often only 2D visualization or missing transformation tools What if I’m not an image processing expert but a clinician or researcher from a different field?
  9. What if we use simple interaction techniques and real-time visual feedback to simplify the registration process? With this in mind, we designed a system that allow a user to import two volumes and create simple 3D visualizations. The user can perform transformations by interacting with the moving volume to register it to the target volume, while our system provides feedback on the current registration result. After each registrationstep, the user can evaluate the results and either adjust the transformations or accept the result to iteratively converge to a satisfactory registration result
  10. Placing landmarks in the intended location can be challenging in 3D. Therefore we introduce two techniques to easily place corresponding landmarks in 3D. If a surface can be visualized for a structure of interest, the user can place landmarks directly on the surface using the surface picker. The arrow indicates the surface normal and is used as a cursor for landmark picking
  11. When the structures of interest can not be clearly visualized as a surface and occlusion is present, the two-step picker can be used. In the first step, the user defines a ray from the camera to the selected point, while in the second step, the intended depth along the ray can be picked. An intensity profile of the values encountered along the ray with a linked cursor aids the user in specifying the correct depth
  12. We introduce simple visualization techniques to quickly visualize structures of interest in 3D Double thresholds and opacity values can be set to make simple 3D visualizations. When the structures of interest are of high intensity a maximum intensity projection can be selected.
  13. We visualize the current registration result using multivolume-rendering techniques in 3D for each of the three potential visualization combinations. By always displaying the target volume in orange and the moving volume in blue, the volumes can be identified. Furthermore, in the MIP+MIP view, these colors sum to white when there is overlap.
  14. The full system is called RegistrationShop. It combines the aforementioned techniques and several more features in a single application. Target volume left in orange Moving volume right in blue Registration result in the center
  15. RegistrationShop is implemented in Python using VTK extended with custom multi-volume rendering features. We rely on Elastix for advanced registration techniques.
  16. Let’s take a look at some of these features in a short demo video.
  17. We evaluated RegistrationShop with four domain experts. RegistrationShop was considered a promising volume registration tool with the potential to make registration more accessible to clinicians and researchers in the future.
  18. We evaluated RegistrationShop with four domain experts. RegistrationShop was considered a promising volume registration tool with the potential to make registration more accessible to clinicians and researchers in the future.
  19. 3D: Appreciated, but need to become familiar with them first 2D is required for detailed inspection of the results Transformation tools: Good manual initialization possible for subsequent steps Two-step landmark picking was considered useful Real-time visualization of intermediate registration result very helpful
  20. There is still room for improvements such as visualizing registration quality, visualizing non-rigid deformation field, experimenting with compositing operations or transfer function definitions, examining clinical potential in follow-up user study.
  21. We presented the RegistrationShop system: a prototype registration software tool that provides users with simple transformation tools combined with visual feedback of the registration result