$ xvr --help

Usage: xvr [OPTIONS] COMMAND [ARGS]...

  A PyTorch package for 2D/3D XRAY to CT/MR registration.

  Provides functionality for rapidly training pose regression models and
  registering clinical data with gradient-based iterative optimization.

Options:
  -v, --version  Show the version and exit.
  -h, --help     Show this message and exit.

Commands:
  train     Train a pose regression model.
  restart   Restart model training from a checkpoint.
  register  Use gradient-based optimization to register XRAY to a CT/MR.
  animate   Animate the trajectory of iterative optimization.
  dcm2nii   Convert a DICOMDIR to a NIfTI file.

$ xvr train --help

Usage: xvr train [OPTIONS]

  Train a pose regression model.

Required options:
  -v, --volpath PATH              A single CT or a directory with multiple volumes for pretraining  [required]
  -o, --outpath PATH              Directory in which to save model weights  [required]

Data options:
  -m, --maskpath PATH             Optional labelmaps corresponding to the CTs passed in `volpath`
  --patch_size TEXT               Optional random crop size (e.g., 'h,w,d'); if None, return entire volume
  --num_workers INTEGER           Number of subprocesses to use in the dataloader  [default: 4]
  --pin_memory                    Copy volumes from the dataloader into CUDA pinned memory before returning

Sampling options:
  --r1 <FLOAT FLOAT>...           Range for primary angle (in degrees)  [required]
  --r2 <FLOAT FLOAT>...           Range for secondary angle (in degrees)  [required]
  --r3 <FLOAT FLOAT>...           Range for tertiary angle (in degrees)  [required]
  --tx <FLOAT FLOAT>...           Range for x-offset (in millimeters)  [required]
  --ty <FLOAT FLOAT>...           Range for y-offset (in millimeters)  [required]
  --tz <FLOAT FLOAT>...           Range for z-offset (in millimeters)  [required]
  --batch_size INTEGER            Number of DRRs per batch  [default: 116]

Renderer options:
  --sdd FLOAT                     Source-to-detector distance (in millimeters)  [required]
  --height INTEGER                DRR height (in pixels)  [required]
  --delx FLOAT                    DRR pixel size (in millimeters / pixel)  [required]
  --renderer [siddon|trilinear]   Rendering equation  [default: trilinear]
  --orientation [AP|PA]           Orientation of CT volumes  [default: AP]
  --reverse_x_axis                Enable to obey radiologic convention (e.g., heart on right)

Model options:
  --model_name TEXT               Name of model to instantiate from the timm library  [default: resnet18]
  --norm_layer TEXT               Normalization layer  [default: groupnorm]
  --pretrained                    Load pretrained ImageNet-1k weights
  --parameterization TEXT         Parameterization of SO(3) for regression  [default: quaternion_adjugate]
  --convention TEXT               If `parameterization='euler_angles'`, specify order  [default: ZXY]
  --unit_conversion_factor FLOAT  Scale factor for translation prediction (e.g., from m to mm)  [default: 1000.0]
  --p_augmentation FLOAT          Base probability of image augmentations during training  [default: 0.333]

Optimizer options:
  --lr FLOAT                      Maximum learning rate  [default: 0.0002]
  --weight_ncc FLOAT              Weight on mNCC loss term  [default: 1.0]
  --weight_geo FLOAT              Weight on geodesic loss term  [default: 0.01]
  --weight_dice FLOAT             Weight on Dice loss term  [default: 1.0]
  --weight_mvc FLOAT              Weight on multiview consistency loss term  [default: 0]
  --n_total_itrs INTEGER          Number of iterations for training the model  [default: 1000000]
  --n_warmup_itrs INTEGER         Number of iterations for warming up the learning rate  [default: 1000]
  --n_grad_accum_itrs INTEGER     Number of iterations for gradient accumulation  [default: 4]
  --n_save_every_itrs INTEGER     Number of iterations before saving a new model checkpoint  [default: 1000]
  --disable_scheduler             Turn off cosine learning rate scheduler

Checkpoint options:
  -c, --ckptpath PATH             Checkpoint of a pretrained pose regressor
  --reuse_optimizer               If ckptpath passed, initialize the previous optimizer's state
  -w, --warp PATH                 SimpleITK transform to warp input CT to the checkpoint's reference frame
  --invert                        Whether to invert the warp or not

Logging options:
  --name TEXT                     WandB run name
  --id TEXT                       WandB run ID (useful when restarting from a checkpoint)
  --project TEXT                  WandB project name  [default: xvr]

$ xvr register model --help

Usage: xvr register model [OPTIONS] XRAY...

  Initialize from a pose regression model.

Required options:
  -c, --ckptpath PATH            Checkpoint of a pretrained pose regressor  [required]
  -v, --volume PATH              Input CT volume (3D image)  [required]
  -o, --outpath PATH             Directory for saving registration results  [required]

Renderer options:
  -m, --mask PATH                Labelmap for the CT volume
  --labels TEXT                  Labels in mask to exclusively render (comma-separated)
  --reverse_x_axis               Enable to obey radiologic convention (e.g., heart on right)
  --renderer [siddon|trilinear]  Renderer equation  [default: trilinear]
  --voxel_shift FLOAT            Position of voxel (top left corner or center)  [default: 0.0]

Preprocessing options:
  --crop INTEGER                 Center crop the X-ray image  [default: 0]
  --subtract_background          Subtract mode X-ray image intensity
  --linearize                    Convert X-ray from exponential to linear form
  --equalize                     Apply histogram equalization to X-rays/DRRs during optimization
  --reducefn TEXT                If DICOM is multiframe, method to extract a single 2D image  [default: max]
  --pattern TEXT                 Pattern rule for glob is XRAY is directory  [default: *.dcm]

Optimizer options:
  --scales TEXT                  Scales of downsampling for multiscale registration (comma-separated)  [default: 8]
  --n_itrs TEXT                  Number of iterations to run at each scale (comma-separated)  [default: 500]
  --parameterization TEXT        Parameterization of SO(3) for regression  [default: euler_angles]
  --convention TEXT              If parameterization is Euler angles, specify order  [default: ZXY]
  --lr_rot FLOAT                 Initial step size for rotational parameters  [default: 0.01]
  --lr_xyz FLOAT                 Initial step size for translational parameters  [default: 1.0]
  --patience INTEGER             Number of itrs without improvement before decreasing the learning rate  [default: 10]
  --threshold FLOAT              Threshold for measuring the new optimum  [default: 0.0001]
  --max_n_plateaus INTEGER       Number of times loss can plateau before moving to next scale  [default: 3]

Logging options:
  --init_only                    Directly return the initial pose estimate (no iterative pose refinement)
  --saveimg                      Save ground truth X-ray and predicted DRRs
  --verbose INTEGER RANGE        Verbosity level for logging  [default: 1; 0<=x<=3]

Miscellaneous options:
  --warp PATH                    SimpleITK transform to warp input CT to a template reference frame
  --invert                       Whether to invert the warp or not

@article{grupp2020automatic,
  title={Automatic annotation of hip anatomy in fluoroscopy for robust and efficient 2D/3D registration},
  author={Grupp, Robert B and Unberath, Mathias and Gao, Cong and Hegeman, Rachel A and Murphy, Ryan J and Alexander, Clayton P and Otake, Yoshito and McArthur, Benjamin A and Armand, Mehran and Taylor, Russell H},
  journal={International journal of computer assisted radiology and surgery},
  volume={15},
  pages={759--769},
  year={2020},
  publisher={Springer}
}

@article{pernus20133d,
  title={3D-2D registration of cerebral angiograms: A method and evaluation on clinical images},
  author={Mitrović, Uroš and Špiclin, Žiga and Likar, Boštjan and Pernuš, Franjo},
  journal={IEEE transactions on medical imaging},
  volume={32},
  number={8},
  pages={1550--1563},
  year={2013},
  publisher={IEEE}
}