Dataset: Segmentation of cortical bone, trabecular bone, and medullary pores from micro-CT images using 2D and 3D deep learning models

Computed tomography (CT) enables rapid imaging of large-scale studies of bone, but those datasets typically require manual segmentation, which is time-consuming and prone to error.  Convolutional neural networks (CNNs) offer an automated solution, achieving superior performance on image data.  Here, we used CNNs to train segmentation models from scratch on 2D and 3D patches from micro-CT scans of otter long bones.  These new models, collectively called BONe (Bone One-shot Network), aimed to be fast and accurate, and we expected enhanced results from 3D training due to better spatial context.  Our results showed that 3D training required substantially more memory.  Contrary to expectations, 2D models performed slightly better than 3D models in labeling details such as thin trabecular bone.  Although lacking in some detail, 3D models appeared to generalize better and predict smoother internal surfaces than 2D models.  However, the massive computational c..., Materials
Limb bones from the North American river otter (Lontra canadensis) were borrowed from four museums — OMNH (SNM): Sam Noble Oklahoma Museum of Natural History (Norman Oklahoma); UAM: University of Alaska Museum of the North (Fairbanks, Alaska); and UF: Florida Museum of Natural History (Gainesville, Florida); UWBM (BM): Burke Museum of Natural History and Culture (Seattle, Washington).  In total, the sample consisted of 38 elements (humerus, radius, ulna, femur, tibia, and fibula) from nine individuals.

Specimen

kV

µA

Filter

Res. (µm)

Provenance

Sex

Side

Element

Group

OMNH 44262

160

312

Copper

49.99

Tennessee

F

R

Humerus

Fitting

 

 

 

 

 

 

 

L

Radius

Fitting

 

 

 

 

 

 

 

L

Ulna

Fitting

OMNH 53994

160

312

Copper

49.99

Tennessee

M

L

Femur

Fitting

 

 

 

 

 

&..., , # Dataset: Segmentation of cortical bone, trabecular bone, and medullary pores from micro-CT images using 2D and 3D deep learning models

[https://doi.org/10.5061/dryad.b2rbnzsq4](https://doi.org/10.5061/dryad.b2rbnzsq4)

Introduction: The following document outlines the structure of data repository.

“_composite.zip” contains separate folders for the raw micro-CT slices and CTP labels used for model fitting.  Because Avizo’s deep learning segmentation module can only load a single pair of data files (i.e., one file consists of the raw slices and the second file consists of the labels), we appended the samples used for fitting into a composite sample.

“_Models.zip” contains the deep learning models for bone-pore and cortical-trabecular-pores segmentation.  Model files are intended for use in the commercial software Avizo/Amira but are compatible with a variety of both open source and commercial software (e.g., TensorFlow or Comet Dragonfly).  Files consist of model weights (HDF5 format...