Project 1 journal

Shreya D'Souza

Click here to see all my wiki deliverables

WEEK 1

25th Jan (5 hours)

  • Created this journal page

  • Read through last year's wiki

    • Looked at journals from last year to understand the expectations for the journal and how in-depth the entries need to be

  • Read through papers about VR and ways in which data can be visualized using VR

WEEK 2

27th Jan (2 hours)

  • Installed Paraview

    • This took longer than I anticipated as I thought I had to download CMake as well, as opposed to just the binaries

  • Read an article about Data Visualization in Virtual Reality

    • Learnt about WebVR, A-Frame and D3

28th Jan (1.5 hours)

  • Played around with different color map settings on Paraview to see how increasing the intensity of certain colours would change what we were viewing

  • Saw how different densities affected what was being shown in the visualization

  • Tried to use Paraview for other datasets as suggested in the slack

30th Jan 2020 (2.5 hours)

  • Looked at different software that I could use for my project

  • I thought SlicerVR was interesting but it seems as though it is only available on Windows -- need to ask if there are other alternatives

  • If not, planning on finding other software that would allow me to visualise datasets that are related to the medical field in some way

  • Alternative: OpenSpace -- use it for NASA's Earth Data


2nd Feb (4 hours)

  • Working on trying to use 3D Slicer, as it is available for OS X

  • Project Ideas:

    1. Using 3D Slicer to visualize the affect of chemotherapy(e.g. IL2) on brain tumours through MRIs

      • Need to find data set -- maybe ask professor mentioned during class?

    2. Using 3D Slicker to take a closer look at MRIs depicting the progression of Alzheimer's

    3. Using OpenSpace to analyse NASAs Earthdata

      • Not sure what part of data I'm most interested in or the application of this

    4. Backup: Look more into CLARITY

  • All in all, I need to find a purpose of my project: we know that chemotherapy attacks cancerous cells, so how can viewing data showing this help?

  • Downloaded 3D Slicer to test how the software works (will create a page for the Wiki)

  • Downloaded Unreal Engine

  • Requested an Oscar Account

WEEK 3

3rd Feb (2 hours)

  • Read a paper outlining how fMRIs can be visualised in VR

    • Learnt more about the science behind fMRI to get an idea of what exactly datasets will be showing

  • Created a page for 3D Slicer


6 Feb (4 hours)

Evaluations



Review by Brandon Woodard:

o The proposed project clearly identifies deliverable additions to our VR Software Wiki

Strongly agree; The project will have new entries in the functionality of 3D slicer and may have new entries on visualization techniques most appropriate for MRI data. The project will also have additions to the wiki that pertain to using the GlioBlastoma Multiforme dataset.

o The proposed project involves previously unavailable Yurt data visualization functionality

Agree; the project involves using the GlioBlatoma Multiforme dataset which has not been visualized in the YURT yet. In order to make this possible the 3D slicer software would have to be used and have to be made functional in the YURT.

o The proposed project involves large data visualization along the lines of the "Data Types" wiki page and identifies the specific data and software that it will use

Agree; The proposed project will involve the large dataset of GlioBlatoma Multimforme dataset which provides 'slices' of the brain that present tumor data. She has identified the dataset to be in '(dot)TCIA' formats.

o The proposed project has a realistic schedule with explicit and measurable milestones at least each week and mostly every class

Neither agree nor disagree; The proposed project has a schedule that correlates with the milestone schedule provided by the course homepage: 2/13, 2/20, 2/25, 3/03, 3/05, 3/10 , but the milestones seemed more general even though she did cover the main points. I think the most important milestones were addressed, but more detailed milestones should be included. I recommend more thought behind how she plans on completing each large milestone and breaking that up into smaller ones.

o The proposed project includes an in-class activity

Agree: Yes the proposed project includes a class activity

o The proposed project has resources available with sufficient documentation

Agree: Not sure how granular the level of detail is with the documentation, but it does exist.

Final Feedback/Recommendations:

  • I recommend that you find a more recent dataset, the dataset that you're using was last updated in 1992. We already spoke about this in class so I know you're on it!

  • I would've liked to hear how you plan on creating the video of brain tumor progression you mentioned in class. This was one of your deliverables. This will just require more research as to how to get this done, maybe its been done before with 3D slicer?


Shreya Project Evaluation by David 02/07/20

o The proposed project clearly identifies deliverable additions to our VR Software Wiki

4 a video will need some wrapping to make it work in the wiki. What will it tell our wiki users? I think that the users want to know what they can do with VR and visaulization as opposed to how tumors change size. Similarly, what will the poster tell them?

o The proposed project involves previously unavailable Yurt data visualization functionality

3 I fear that your "work on yurt portability" needs to be fleshed out quite a bit. How will you do that? That could take many weeks of work, which I don't see the tie for. I would suggest moving that element forward in your schedule. If you can better determine before Tuesday the steps you would need to do to port, build that in to your schedule. Which of the paths that Ben described will get 3DSlicer to work in the Yurt?

o The proposed project involves large data visualization along the lines of the "Data Types" wiki page and identifies the specific data and software that it will use

4 Your data plan sounds pretty well fleshed out. I think it's probably ok to work with the older data. Making 3DSlicer work in the Yurt will be a tremendous accomplishment, and illustrating it even with older data will capture that.

o The proposed project has a realistic schedule with explicit and measurable milestones at least each week and mostly every class

2-3 My concerns about Yurt porting repeat...

o The proposed project includes an in-class activity

4 Something seems likely to emerge :-). But if there isn't anything that would help you create a deliverable, that could be ok.

o The proposed project has resources available with sufficient documentation

4 My concerns about Yurt porting repeat... the data parts seem pretty good, although the understanding of GBMs might not be quite up to date.

o Also add any additional thoughts guided by the questions in the homework wiki entry.

My understanding is that cancer treatment for GBM's has not improved much in many year, but the understanding of the disease has. I think that getting more recent data is probably possible. Dr. Richard Gilbert (Richard_Gilbert@brown.edu) has a particular interest in this. You could reach out to him for ideas or possibly data. You could also look in google scholar for research on this involving imaging data. He will likely be very excited, but he may also have more ideas than can be accommodated, so keep that in mind :-)

All that said, MRI and CT imaging has not changed a ton, so the data that you have found may illustrate the value of your visualization approach.

I don't have permission for the google drive link on your project page.

I think it is unlikely that you will find tumor progression data. It's very hard to keep imaging patients over time.



Shreya Project Evaluation by Ross 02/07/20

o The proposed project clearly identifies deliverable additions to our VR Software Wiki: 4

Good, concrete deliverables. Would the Slicer 3D tutorial chronicle how to visualize MRI / CT data in the YURT? If not, it would be beneficial if there was a YURT related deliverable.

o The proposed project involves previously unavailable Yurt data visualization functionality: 5

Currently, viewing volumetric data in the YURT is very difficult, so being able to visualize MRI / CT would be incredibly useful. However, I agree with data that the task will be non-trivial and will require detailed planning.

o The proposed project involves large data visualization along the lines of the "Data Types" wiki page and identifies the specific data and software that it will use: 4

I'm not sure if the "oldness" of the scans will impact the results; over the past twenty years, I think that MRI resolution has improved quite a bit, so that is something to consider. In addition, given that you'd like to highlight where a tumor resides within the brain, I'd strongly consider looking at the BRATS dataset, which I've linked in the wiki. This data includes MRI scans where the tumors have already been segmented by a trained professional, so this may help with you task. In addition, BRATS is the benchmark dataset for brain tumor segmentation.

o The proposed project has a realistic schedule with explicit and measurable milestones at least each week and mostly every class: 5

Milestones are concrete and detailed.

o The proposed project includes an in-class activity: 4

Good overview of in-class activities; would be beneficial to add more details for project presentation.

o The proposed project has resources available with sufficient documentation: 4

Getting Slicer to work in the YURT might be a difficult task, so I'd think about where you'd turn next if porting Slicer seems infeasible. Overall, good work though!


Evaluation by Giuse 2/8/20

o The proposed project clearly identifies deliverable additions to our VR Software Wiki

4 Deliverable additions clearly identified, but the wiki should also have text/table other forms of info available

o The proposed project involves previously unavailable Yurt data visualization functionality

5 This new YURT data would be awesome

o The proposed project involves large data visualization along the lines of the "Data Types" wiki page and identifies the specific data and software that it will use

5 Specific and large data visualization explained

o The proposed project has a realistic schedule with explicit and measurable milestones at least each week and mostly every class

5 Looks great

o The proposed project includes an in-class activity

5 Activity is related and useful

o The proposed project has resources available with sufficient documentation

4 Medical data visualization seems supportable, but I agree with David and Ross on the feasibility of Slicer.

9 Feb (2 hours)

  • Downloaded several data sets from The Cancer Imaging Archive and visualized them to see what looks best in terms of how clear it is and the size of the tumour

10 Feb (2 hours)

  • Created my presentation for class, decided what I was going to say

Self Evaluation

The proposed project clearly identifies deliverable additions to our VR Software Wiki

5: Will be creating a tutorial and a video

The proposed project involves previously unavailable Yurt data visualization functionality

3: I have determined how I can view the scans in VR, but I'm not sure how I will be able to go from that to porting it to the YURT.


The proposed project involves large data visualization along the lines of the "Data Types" wiki page and identifies the specific data and software that it will use

5: Using huge data sets from TCIA and using 3D Slicer

The proposed project has a realistic schedule with explicit and measurable milestones at least each week and mostly every class

4: Need to expand upon YURT portability

The proposed project includes an in-class activity

5

The proposed project has resources available with sufficient documentation

4: Again need to figure out portability

WEEK 4

12 Feb (4 hours)

  • Downloaded all the BRAATS data

    • Tumours are already coloured, which removes a step in my process

    • Generated 3D MRI models

Vid 3D MRI.mov

13 Feb (3 hours)

  • Got a 3D rendering of the brain (not the tumour)

14 Feb (3 hours)

  • Created a 3D model of the brain with the tumour

  • The BRAATS data already has the tumour labelled, so fortunately I didn't have to manually colour the tumour

Screen Recording 2020-02-14 at 12.31.32 PM.mov

WEEK 5

15 Feb (3.5 hours)

18 Feb (2.5 hours)

  • Created and sent presentation documenting my progress on my project

  • Need to determine how to port to YURT - Paraview is currently looking like the most suitable option because my models seem to look like polygonal models with transparency

  • Read this: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4181673/

Possible pipeline? (From source - need to flesh this out)

Self Evaluation (Journal)

Criteria:

Journal activities are explicitly and clearly related to course deliverables - 4

deliverables are described and attributed in wiki - 4

report states total amount of time - 5

total time is appropriate - 5

Anessa: Self-evaluation is an accurate representation of work!

Evaluation By Brandon Woodard

Journal activities are explicitly and clearly related to course deliverables - 5

deliverables are described and attributed in wiki - 5

report states total amount of time - 5

Feedback: Could add total time count, but I don't know if this is necessary

total time is appropriate - 5

Feedback: Time is accurate and realistic. I did not think there was any over estimation or under estimation as I went through this.

Evaluation of self-evaluation: Overall this journal is well organized and thought out. I just recommend making all google drive items public and I think you have an opportunity to document the BRAATS data set on the wiki. This would be a good addition. I think you were a little hard on yourself when it came to deliverables only because your time was spent working on your project, so you can add your wiki additions or start adding as you learn about new content.

22 Feb (3 hours)

  • Generated a model using Paraview (exported segmentations as vtk files)

  • Unsure how to reduce opacity of brain tissue - currently using "wireframe" visualisation

  • Need to work on porting the model to HMDs -> 3D Slicer has a VR extension but it only works on Windows/maybe Linux

Screen Recording 2020-02-22 at 2.40.54 PM.mov

23 Feb (1 hour)

  • Planned class activity

    • Will be porting 3D models to HMDs -- need to try this out on Windows computers first.

    • Updated course timeline

WEEK 6

26 Feb (2 hours)

  • Finished 3D model in Paraview and sent script to Ross for viewing in the YURT on Thursday

28 Feb (4 hours)

  • Made the in-class activity

  • Went to the Visual Computing Lab and got a model to show up in Slicer VR -- need to use the middle computer in the glass room

29 Feb (4 hours)

  • Got the activity to work on the HTC Vives

  • Wrote up the activity for class

  • Parts of the brain can be moved around with the controllers

WEEK 7

I had many deadlines this week so was not able to put as much time into the class as I would have liked

2 March (2 hours)

  • Went back to the VC Lab to document the process of getting the model to show up on the HTC Vives.

3 March (2 hours)

  • Tried to find a way to make the brain tissue transparent in Paraview

5 March (1 hour)

  • Downloaded XCode and attempted Anessa's Google Cardboard Tutorial (I was not able to complete it in class)

8 March (3 hours)

  • Prepared two models for porting to the YURT, visited the YURT and visualized them

    • Was not able to log into the computer in the YURT

WEEK 8

Spent a lot of the week packing to go back home, apologies for spending fewer than 10 hours!

9 March (6 hours)

  • Prepared for the class presentation

  • Thought of some ideas for the second project