Code Monkey home page Code Monkey logo

afids-regrf's Introduction

AFIDs

An open framework for evaluating correspondence in brain images and teaching neuroanatomy using anatomical fiducial placement

AFIDs

Preprint: https://www.biorxiv.org/content/10.1101/460675v2

Manuscript: http://dx.doi.org/10.1002/hbm.24693

Documentation: https://afids.readthedocs.io/en/latest/

Join our channel on

Introduction

  • The AFID protocol is an anatomical fiducial placement protocol that has been validated and used for teaching at a number of local and BHG-related events including https://github.com/BrainhackWestern/BrainhackWestern.github.io/wiki/Tutorials.
  • AFID placement is reproducible, not overtly manually intensive (20-40 minutes once trained), and more sensitive to local registration errors than standard voxel overlap measures.
  • This protocol and study framework leverages open resources and tools, and has been developed with full transparency in mind so that others may freely use, adopt, and modify.
  • 60+ raters trained to date.

afids

References and Resources

Preprint

  • Lau JC, Parrent AG, Demarco J, Gupta G, Park PJ, Ferko K, Khan AR, Peters TM. A framework for evaluating correspondence between brain images using anatomical fiducials. bioRxiv. 2018. [ref]

Conference Abstracts

  • Lau JC, Parrent AG, Demarco J, Gupta G, Park PJ, Ferko K, Khan AR, Peters TM. AFIDs: an open framework for evaluating correspondence between magnetic resonance images of the human brain using fiducial placement. F1000 Research. Demo presented at INCF NeuroInformatics in Montreal, QC, Canada. 2018. [ref]

Open Datasets

  • Agile12v2016: Lau JC, MacDougall KW, Arango MF, Peters TM, Parrent AG, Khan AR: Ultra-High Field Template-Assisted Target Selection for Deep Brain Stimulation Surgery. World Neurosurg 103:531–537, 2017. [download] [ref]
  • Colin27: Holmes CJ, Hoge R, Collins L, Woods R, Toga AW, Evans AC: Enhancement of MR Images Using Registration for Signal Averaging. J Comput Assist Tomogr 22:324–333, 1998. [download] [ref]
  • MNI152NLin2009bAsym: Fonov V, Evans AC, Botteron K, Almli RR, McKinstry RC, Collins LL, et al: Unbiased average age-appropriate atlases for pediatric studies. Neuroimage 54:313–327, 2011. [download] [ref]
  • OASIS1: Marcus DS, Fotenos AF, Csernansky JG, Morris JC, Buckner RL: Open Access Series of Imaging Studies: Longitudinal MRI Data in Nondemented and Demented Older Adults. J Cogn Neurosci 22:2677–2684, 2010. [download] [ref]

Software

  • 3D Slicer: Fedorov A, Beichel R, Kalpathy-Cramer J, Finet J, Fillion-Robin JC, Pujol S, et al: 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn Reson Imaging 30:1323–1341, 2012. [download] [ref]

Historical References

  • Talairach1957: Talairach J, David M, Tournoux P, Corredor H, Kvasina T: Atlas d’anatomie Stéréotaxique. Repérage Radiologique Indirect Des Noyaux Gris Centraux Des Régions Mésencephalosousoptique et Hypothalamique de l’homme. Paris, France: Masson & Cie, 1957
  • Talairach1988: Talairach J, Tournoux P: Co-Planar Stereotaxic Atlas of the Human Brain: 3-D Proportional System: An Approach to Cerebral Imaging. ed 1, Thieme, 1988.

Other Resources

afids-regrf's People

Contributors

ataha24 avatar kaitj avatar tkkuehn avatar

Watchers

 avatar  avatar  avatar

Forkers

tkkuehn

afids-regrf's Issues

Change model output format to bundle relevant info for model application.

To run the model application correctly, right now you need to both know and accurately input:

  • The feature offsets used to train the model (via a feature_offsets.npz)
  • The padding, size, and sampling rate (directly via CLI parameters)

I think it would be a big QoL improvement to bundle the feature offsets file and other three parameters (via a YAML or TOML file of some sort) with the model outputs. We could then zip them up with the model outputs (with a custom extension like .afidsregrf), so the end user just has to provide the one file to apply.py to correctly apply their model.

restrict sampling of voxels to be within image boundaries

Currently the feature extraction (i.e., offsets for haar features) are insensitive to image boundaries. When the choen voxel to compute the samples is close to the endge of the image, the code will break. The solution may be to restrict offset sampling to keep the voxels within the boundaries of the image and simply discard the ones that are not. Attached is an example of how the code fails at inference.

Screenshot 2023-03-28 at 9 27 37 PM

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.