File size: 1,723 Bytes
d9a816e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 | {
"model_type": "vesselboost-unet3d",
"framework": "pytorch",
"task": "image-segmentation",
"architecture": {
"class_name": "models.unet_3d.Unet",
"spatial_dimensions": 3,
"input_channels": 1,
"output_channels": 1,
"base_filters": 16,
"encoder_stages": 4,
"decoder_stages": 4,
"convolution_kernel_size": [3, 3, 3],
"output_kernel_size": [1, 1, 1],
"normalization": "batch_norm_3d",
"activation": "relu",
"output_activation": "sigmoid_at_inference"
},
"input": {
"format": "NIfTI",
"contrasts": [
"TOF-MRA",
"T2*-weighted MRI (experimental checkpoint only)"
],
"patch_size": [64, 64, 64],
"default_patch_stride": [64, 64, 64]
},
"preprocessing": {
"resize_target": "each spatial dimension is at least 64 and a multiple of 64",
"resize_interpolation": "nearest_neighbor",
"intensity_transform": {
"name": "z_score_standardization",
"scope": "whole_volume",
"formula": "(x - mean(x)) / std(x)",
"constant_volume_result": "zeros"
},
"optional_operations": [
"N4 bias-field correction",
"denoising",
"brain extraction with separately distributed SynthStrip weights"
]
},
"postprocessing": {
"probability_threshold": 0.1,
"connected_component_minimum_voxels": 10,
"connected_component_connectivity": 26,
"prediction_resize_interpolation": "nearest_neighbor"
},
"checkpoint_format": "PyTorch state_dict ZIP serialization",
"compatible_vesselboost": {
"version": "2.0.2",
"git_tag": "v2.0.2",
"git_commit": "1504b00c91777d5e2c271c1cab7f500078f08c69",
"source_url": "https://github.com/KMarshallX/VesselBoost/tree/v2.0.2"
}
}
|