ANTSJOINTTENSORFUSION(1)

✨ ANTSJOINTTENSORFUSION(1) ✨ — User Commands 💖

NAME

Okay so the manual name is antsJointTensorFusion — part of the ANTS registration suite 🧚

---

DESCRIPTION

💅 COMMAND:

antsJointTensorFusion

Bestie, antsJointTensorFusion is an image fusion algorithm that literally slayed segmentation challenges at MICCAI 2012 and MICCAI 2013 👑. Hongzhi Wang and Paul Yushkevich made the original label fusion framework, and then Brian Avants extended it so it could handle intensities too 🌈. This implementation is based on Paul's original ITK-style version and Brian's ANTsR version, so it's giving legacy energy ✨.

If you want the receipts, the references are: 1) H. Wang, J. W. Suh, S. Das, J. Pluta, C. Craige, P. Yushkevich, Multi-atlas segmentation with joint label fusion, IEEE Trans. on Pattern Analysis and Machine Intelligence, 35(3), 611-623, 2013. 📚 and 2) H. Wang and P. A. Yushkevich, Multi-atlas segmentation with joint label fusion and corrective learning--an open source implementation, Front. Neuroinform., 2013. 🦋

✨ OPTIONS ✨

• -d, --image-dimensionality 2/3/4 — Forces the image to be treated as a specific-dimensional image 🎀. If you don't set it, the program tries to figure out the dimensionality from the input image on its own 🙈.
• -t, --target-image targetImage [targetImageModality0,targetImageModality1,...,targetImageModalityN] — The target image (or multimodal target images) that's assumed to be aligned to a common image domain 🫶.
• -g, --atlas-image atlasImage [atlasImageModality0,atlasImageModality1,...,atlasImageModalityN] — The atlas image (or multimodal atlas images) that's assumed to be aligned to a common image domain 🫶.
• -l, --atlas-segmentation atlasSegmentation — The atlas segmentation images. For label fusion, the number of segmentations you give has to match the number of atlas image sets, no exceptions 💀.
• -a, --alpha 0.1 — A regularization term added to matrix Mx before the inverse gets calculated 🧪. Default = 0.1.
• -b, --beta 2.0 — The exponent used to map intensity difference to the joint error 📈. Default = 2.0.
• -r, --retain-label-posterior-images (0)/1 — Keeps the label posterior probability images 💜. This one needs atlas segmentations to be specified first. Default = false.
• -f, --retain-atlas-voting-images (0)/1 — Keeps the atlas voting images 🗳️. Default = false.
• -c, --constrain-nonnegative (0)/1 — Constrains the solution to non-negative weights 🍓.
• -u, --log-euclidean (0)/1 — Uses log Euclidean space for tensor math 🌸.
• -p, --patch-radius 2 2x2x2 — The patch radius for similarity measures 🔍. Default = 2x2x2.
• -m, --patch-metric (PC)/MSQ — The metric used to figure out which neighborhood patch is the most similar 💅. You can pick Pearson's correlation (PC) or mean squares (MSQ). Default = PC (Pearson correlation).
• -s, --search-radius 3 3x3x3 — The search radius for similarity measures 🛰️. Default = 3x3x3.
• -e, --exclusion-image label[exclusionImage] — Lets you specify an exclusion region for the given label 🚫.
• -x, --mask-image maskImageFilename — If you give it a mask image, fusion only happens inside the mask region 🎭.
• -o, --output labelFusionImage intensityFusionImageFileNameFormat [labelFusionImage,intensityFusionImageFileNameFormat,<labelPosteriorProbabilityImageFileNameFormat>,<atlasVotingWeightImageFileNameFormat>] — This is your output: the intensity and/or label fusion image 🎉. You can also grab the optional extras, which are the label posterior probability images and the atlas voting weight images 🥰.
• --version — Shows you the version info 🔖.
• -v, --verbose (0)/1 — Verbose output, so the program spills all the tea 🍵.
• -h — Prints the help menu (short version) 🆘.
• --help — Prints the help menu. <VALUES>: 1 📖

---

antsJointTensorFusion 2.6.4+dfsg, December 2025 — bestie, that's the whole page, and you're officially the Unix girlie of the group chat 🦋💖 periodt! 💅✨