update and debug

260_gt_nnUNetResEncUNetLPlans.json

@@ -0,0 +1,521 @@
+{
+    "dataset_name": "Dataset261_synthrad2025_task1_CT_AB_pre_v2r_stitched_masked",
+    "plans_name": "nnUNetResEncUNetLPlans",
+    "original_median_spacing_after_transp": [
+        3.0,
+        1.0,
+        1.0
+    ],
+    "original_median_shape_after_transp": [
+        99,
+        442,
+        465
+    ],
+    "image_reader_writer": "SimpleITKIO",
+    "transpose_forward": [
+        0,
+        1,
+        2
+    ],
+    "transpose_backward": [
+        0,
+        1,
+        2
+    ],
+    "configurations": {
+        "2d": {
+            "data_identifier": "nnUNetPlans_2d",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 13,
+            "patch_size": [
+                448,
+                512
+            ],
+            "median_image_size_in_voxels": [
+                442.0,
+                465.0
+            ],
+            "spacing": [
+                1.0,
+                1.0
+            ],
+            "normalization_schemes": [
+                "CTNormalizationClippingSynthrad2025"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 7,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        512,
+                        512,
+                        512
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv2d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm2d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": true
+        },
+        "3d_lowres": {
+            "data_identifier": "nnUNetResEncUNetLPlans_3d_lowres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                64,
+                192,
+                192
+            ],
+            "median_image_size_in_voxels": [
+                99,
+                292,
+                307
+            ],
+            "spacing": [
+                3.0,
+                1.512589724855112,
+                1.512589724855112
+            ],
+            "normalization_schemes": [
+                "CTNormalizationClippingSynthrad2025"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 6,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        320,
+                        320
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv3d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm3d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": false,
+            "next_stage": "3d_cascade_fullres"
+        },
+        "3d_fullres": {
+            "data_identifier": "nnUNetPlans_3d_fullres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                40,
+                192,
+                224
+            ],
+            "median_image_size_in_voxels": [
+                99.0,
+                442.0,
+                465.0
+            ],
+            "spacing": [
+                3.0,
+                1.0,
+                1.0
+            ],
+            "normalization_schemes": [
+                "CTNormalizationClippingSynthrad2025"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 6,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        320,
+                        320
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv3d",
+                    "kernel_sizes": [
+                        [
+                            1,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm3d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": true
+        },
+        "3d_cascade_fullres": {
+            "inherits_from": "3d_fullres",
+            "previous_stage": "3d_lowres"
+        }
+    },
+    "experiment_planner_used": "nnUNetPlannerResEncL",
+    "label_manager": "LabelManager",
+    "foreground_intensity_properties_per_channel": {
+        "0": {
+            "max": 3958.0,
+            "mean": -226.3246612548828,
+            "median": -81.0,
+            "min": -1502.0,
+            "percentile_00_5": -1024.0,
+            "percentile_99_5": 566.0,
+            "std": 396.30517578125
+        }
+    }
+}

262_gt_nnUNetResEncUNetLPlans.json

@@ -0,0 +1,356 @@
+{
+    "dataset_name": "Dataset263_synthrad2025_task1_CT_HN_pre_v2r_stitched_masked",
+    "plans_name": "nnUNetResEncUNetLPlans",
+    "original_median_spacing_after_transp": [
+        3.0,
+        1.0,
+        1.0
+    ],
+    "original_median_shape_after_transp": [
+        89,
+        296,
+        279
+    ],
+    "image_reader_writer": "SimpleITKIO",
+    "transpose_forward": [
+        0,
+        1,
+        2
+    ],
+    "transpose_backward": [
+        0,
+        1,
+        2
+    ],
+    "configurations": {
+        "2d": {
+            "data_identifier": "nnUNetPlans_2d",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 30,
+            "patch_size": [
+                320,
+                320
+            ],
+            "median_image_size_in_voxels": [
+                296.0,
+                279.0
+            ],
+            "spacing": [
+                1.0,
+                1.0
+            ],
+            "normalization_schemes": [
+                "CTNormalizationClippingSynthrad2025"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 7,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        512,
+                        512,
+                        512
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv2d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm2d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": true
+        },
+        "3d_fullres": {
+            "data_identifier": "nnUNetPlans_3d_fullres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                56,
+                192,
+                160
+            ],
+            "median_image_size_in_voxels": [
+                89.0,
+                296.0,
+                279.0
+            ],
+            "spacing": [
+                3.0,
+                1.0,
+                1.0
+            ],
+            "normalization_schemes": [
+                "CTNormalizationClippingSynthrad2025"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 6,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        320,
+                        320
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv3d",
+                    "kernel_sizes": [
+                        [
+                            1,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm3d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": false
+        }
+    },
+    "experiment_planner_used": "nnUNetPlannerResEncL",
+    "label_manager": "LabelManager",
+    "foreground_intensity_properties_per_channel": {
+        "0": {
+            "max": 3940.0,
+            "mean": -137.8296356201172,
+            "median": 12.0,
+            "min": -1708.0,
+            "percentile_00_5": -1018.0,
+            "percentile_99_5": 1349.0,
+            "std": 482.0824279785156
+        }
+    }
+}

264_gt_nnUNetResEncUNetLPlans.json

@@ -0,0 +1,521 @@
+{
+    "dataset_name": "Dataset265_synthrad2025_task1_CT_TH_pre_v2r_stitched_masked",
+    "plans_name": "nnUNetResEncUNetLPlans",
+    "original_median_spacing_after_transp": [
+        3.0,
+        1.0,
+        1.0
+    ],
+    "original_median_shape_after_transp": [
+        108,
+        476,
+        536
+    ],
+    "image_reader_writer": "SimpleITKIO",
+    "transpose_forward": [
+        0,
+        1,
+        2
+    ],
+    "transpose_backward": [
+        0,
+        1,
+        2
+    ],
+    "configurations": {
+        "2d": {
+            "data_identifier": "nnUNetPlans_2d",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 13,
+            "patch_size": [
+                448,
+                512
+            ],
+            "median_image_size_in_voxels": [
+                476.0,
+                536.0
+            ],
+            "spacing": [
+                1.0,
+                1.0
+            ],
+            "normalization_schemes": [
+                "CTNormalizationClippingSynthrad2025"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 7,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        512,
+                        512,
+                        512
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv2d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm2d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": true
+        },
+        "3d_lowres": {
+            "data_identifier": "nnUNetResEncUNetLPlans_3d_lowres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                64,
+                192,
+                192
+            ],
+            "median_image_size_in_voxels": [
+                98,
+                288,
+                324
+            ],
+            "spacing": [
+                3.278181,
+                1.652847632271752,
+                1.652847632271752
+            ],
+            "normalization_schemes": [
+                "CTNormalizationClippingSynthrad2025"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 6,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        320,
+                        320
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv3d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm3d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": false,
+            "next_stage": "3d_cascade_fullres"
+        },
+        "3d_fullres": {
+            "data_identifier": "nnUNetPlans_3d_fullres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                40,
+                192,
+                224
+            ],
+            "median_image_size_in_voxels": [
+                107.5,
+                476.0,
+                536.0
+            ],
+            "spacing": [
+                3.0,
+                1.0,
+                1.0
+            ],
+            "normalization_schemes": [
+                "CTNormalizationClippingSynthrad2025"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 6,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        320,
+                        320
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv3d",
+                    "kernel_sizes": [
+                        [
+                            1,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm3d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": true
+        },
+        "3d_cascade_fullres": {
+            "inherits_from": "3d_fullres",
+            "previous_stage": "3d_lowres"
+        }
+    },
+    "experiment_planner_used": "nnUNetPlannerResEncL",
+    "label_manager": "LabelManager",
+    "foreground_intensity_properties_per_channel": {
+        "0": {
+            "max": 3867.0,
+            "mean": -282.4960632324219,
+            "median": -94.0,
+            "min": -1607.0,
+            "percentile_00_5": -1024.0,
+            "percentile_99_5": 630.0,
+            "std": 427.12237548828125
+        }
+    }
+}

app.py

@@ -3,7 +3,7 @@ import os
 from huggingface_hub import snapshot_download
 
 from PIL import Image, ImageDraw
-HF_REPO = "aehrc/Synthrad2025"   
+HF_REPO = "Synthard2025KoalAI/synthrad2025_task1"   
 LOCAL_WEIGHTS_DIR = os.path.abspath("weights/task1")  
 
 token = os.getenv("HF_TOKEN")  
@@ -19,7 +19,7 @@ repo_dir = snapshot_download(
     token=token,
 )
 print(repo_dir)
-os.environ["nnUNet_results"] = repo_dir              
+os.environ["nnUNet_results"] = repo_dir        
 os.environ.setdefault("nnUNet_raw", "./nnunet_raw")
 os.environ.setdefault("nnUNet_preprocessed", "./nnunet_preprocessed")
 os.environ["OPENBLAS_NUM_THREADS"] = "1"
@@ -29,13 +29,13 @@ import numpy as np
 import SimpleITK as sitk
 import io
 
-from process import SynthradAlgorithm  
+from process import SynthradAlgorithm2  
 
 st.set_page_config(page_title="SynthRad (nnUNetv2) Demo", layout="wide")
 st.title("SynthRad (MRI/CBCT + Mask → synthetic CT) — Streamlit Demo")
 
 if "algo" not in st.session_state:
-    st.session_state.algo = SynthradAlgorithm()
+    st.session_state.algo = SynthradAlgorithm2()
 if "synth_ct" not in st.session_state:
     st.session_state.synth_ct = None
 if "orig_meta" not in st.session_state:
@@ -76,17 +76,17 @@ SAMPLE_MAP = {
     "Abdomen (sample)": {
         "region": "Abdomen",
         "mri":  os.path.join(repo_dir,  "Abdomen",  "cbct.mha"),
-        "mask": os.path.join(repo_dir,"Abdomen",  "mask.mha"),
+        "mask": os.path.join(repo_dir,"Abdomen",  "mask2.mha"),
     },
     "Head and Neck (sample)": {
         "region": "Head and Neck",
         "mri":  os.path.join(repo_dir,  "Head and Neck", "cbct.mha"),
-        "mask": os.path.join(repo_dir,  "Head and Neck", "mask.mha"),
+        "mask": os.path.join(repo_dir,  "Head and Neck", "mask2.mha"),
     },
     "Thorax (sample)": {
         "region": "Thorax",
         "mri":  os.path.join(repo_dir,  "Thorax",   "cbct.mha"),
-        "mask": os.path.join(repo_dir,  "Thorax",   "mask.mha"),
+        "mask": os.path.join(repo_dir,  "Thorax",   "mask2.mha"),
     },
 }
 c1, c2, c3 = st.columns([2, 2, 1])

app_2.py

@@ -0,0 +1,283 @@
+# app.py
+import os
+import io
+import tempfile
+import zipfile
+import numpy as np
+import SimpleITK as sitk
+import streamlit as st
+
+from PIL import Image, ImageDraw
+from huggingface_hub import snapshot_download
+
+# =========================
+# 配置：两个任务的模型仓库 & 本地路径
+# =========================
+# 你可以将两个任务分别指向不同的 HF repo；如果都在同一个，也可以都填同一个。
+HF_REPOS = {
+    "Task 1 (MR → CT)": "aehrc/Synthrad2025",
+    "Task 2 (CBCT → CT)": "aehrc/Synthrad2025",  # 如有专门CBCT→CT的repo可在此替换
+}
+LOCAL_WEIGHTS_DIRS = {
+    "Task 1 (MR → CT)": os.path.abspath("weights/task1"),
+    "Task 2 (CBCT → CT)": os.path.abspath("weights/task2"),
+}
+
+# 环境设置（Token）
+token = os.getenv("HF_TOKEN")
+if token is None:
+    print("[Warn] HF_TOKEN not set. If the model repo is private, set it in Settings → Variables and secrets.")
+
+# 先下载两个任务的权重（如需按需下载，可在选择任务后再下载）
+REPO_DIRS = {}
+for task_name, repo in HF_REPOS.items():
+    repo_dir = snapshot_download(
+        repo_id=repo,
+        repo_type="model",
+        local_dir=LOCAL_WEIGHTS_DIRS[task_name],
+        local_dir_use_symlinks=False,
+        token=token,
+    )
+    REPO_DIRS[task_name] = repo_dir
+
+# nnUNet 环境变量（指向“当前任务”的 results，会在用户切换任务时动态更新）
+os.environ.setdefault("nnUNet_raw", "./nnunet_raw")
+os.environ.setdefault("nnUNet_preprocessed", "./nnunet_preprocessed")
+os.environ["OPENBLAS_NUM_THREADS"] = "1"
+
+# 从 process.py 导入两个任务的算法类
+# 确保你在 process.py 中定义了 SynthradAlgorithm1（MR→CT）和 SynthradAlgorithm2（CBCT→CT）
+from process import  SynthradAlgorithm2
+
+from process_1 import SynthradAlgorithm1
+
+# =========================
+# Streamlit UI
+# =========================
+st.set_page_config(page_title="SynthRad (nnUNetv2) Demo", layout="wide")
+st.title("SynthRad — 双任务演示（MRI/CBCT + Mask → synthetic CT）")
+
+# 任务选择
+TASKS = ["Task 1 (MR → CT)", "Task 2 (CBCT → CT)"]
+task = st.radio("选择任务", TASKS, index=0, horizontal=True)
+
+# 根据任务设置标题/提示
+if task == "Task 1 (MR → CT)":
+    vol_label = "MRI volume (.nii/.nii.gz/.mha)"
+else:
+    vol_label = "CBCT volume (.nii/.nii.gz/.mha)"
+
+# 动态切换 nnUNet 的 results（不同任务使用不同 results 目录）
+os.environ["nnUNet_results"] = REPO_DIRS[task]
+
+# session_state 初始化
+if "algos" not in st.session_state:
+    st.session_state.algos = {}
+if "synth_ct" not in st.session_state:
+    st.session_state.synth_ct = None
+if "orig_meta" not in st.session_state:
+    st.session_state.orig_meta = None
+if "vol_np" not in st.session_state:
+    st.session_state.vol_np = None
+if "input_vol" not in st.session_state:
+    st.session_state.input_vol = None
+if "input_mask" not in st.session_state:
+    st.session_state.input_mask = None
+
+# 懒加载对应任务的算法实例
+def get_algo(task_name: str):
+    if task_name not in st.session_state.algos:
+        if task_name == "Task 1 (MR → CT)":
+            st.session_state.algos[task_name] = SynthradAlgorithm1()
+        else:
+            st.session_state.algos[task_name] = SynthradAlgorithm2()
+    return st.session_state.algos[task_name]
+
+algo = get_algo(task)
+
+st.subheader("Input")
+src = st.radio("Source", ["Sample", "Upload"], index=0, horizontal=True)
+
+# =========================
+# 样例映射（两任务可共用同一份样例，也可按需区分）
+# 这里假设 repo_dir 下有如下结构：
+#   repo_dir/Abdomen/{cbct.mha, mask.mha} 或 {mri.mha, mask.mha}
+#   repo_dir/Head and Neck/{cbct.mha or mri.mha, mask.mha}
+#   repo_dir/Thorax/{cbct.mha or mri.mha, mask.mha}
+# 如果你的文件名不同，请按需调整。
+# =========================
+
+def build_sample_map(task_name: str):
+    repo_dir = REPO_DIRS[task_name]
+    if task_name == "Task 1 (MR → CT)":
+        vol_key = "mri"
+        vol_fname = "mri.mha"    # 如果你的样例文件名不是 mri.mha，请改成实际名称
+        mask_fname = "mask1.mha"    # 如果你的样例文件名不是 mri.mha，请改成实际名称
+    else:
+        vol_key = "cbct"
+        vol_fname = "cbct.mha"   # 如果你的样例文件名不是 cbct.mha，请改成实际名称
+        mask_fname = "mask2.mha"    # 如果你的样例文件名不是 mri.mha，请改成实际名称
+    sample_map = {
+        "Abdomen (sample)": {
+            "region": "Abdomen",
+            "vol":  os.path.join(repo_dir, "Abdomen",  vol_fname),
+            "mask": os.path.join(repo_dir, "Abdomen", mask_fname),
+        },
+        "Head and Neck (sample)": {
+            "region": "Head and Neck",
+            "vol":  os.path.join(repo_dir, "Head and Neck", vol_fname),
+            "mask": os.path.join(repo_dir, "Head and Neck", mask_fname),
+        },
+        "Thorax (sample)": {
+            "region": "Thorax",
+            "vol":  os.path.join(repo_dir, "Thorax", vol_fname),
+            "mask": os.path.join(repo_dir, "Thorax", mask_fname),
+        },
+    }
+    return sample_map
+
+SAMPLE_MAP = build_sample_map(task)
+
+# =========================
+# 小工具函数
+# =========================
+def _download_sitk_image(img: sitk.Image, file_name: str, label: str):
+    with tempfile.NamedTemporaryFile(suffix=".nii.gz", delete=False) as tmp:
+        sitk.WriteImage(img, tmp.name)
+        tmp_path = tmp.name
+    with open(tmp_path, "rb") as f:
+        st.download_button(
+            label=label,
+            data=f.read(),
+            file_name=file_name,
+            mime="application/octet-stream",
+        )
+    try:
+        os.remove(tmp_path)
+    except Exception:
+        pass
+
+def _read_sitk_from_uploaded(f):
+    suffix = ".nii.gz" if f.name.endswith(".nii.gz") else os.path.splitext(f.name)[1]
+    bio = io.BytesIO(f.read())
+    with tempfile.NamedTemporaryFile(suffix=suffix, delete=False) as tmp:
+        tmp.write(bio.getvalue()); tmp.flush(); path = tmp.name
+    img = sitk.ReadImage(path)
+    try:
+        os.remove(path)
+    except Exception:
+        pass
+    return img
+
+def _read_sitk_from_path(path):
+    if not os.path.exists(path):
+        st.error(f"Sample file missing: {path}")
+        st.stop()
+    return sitk.ReadImage(path)
+
+def _norm2u8(slice2d):
+    s = slice2d.astype(np.float32)
+    s = (s - np.percentile(s, 1)) / (np.percentile(s, 99) - np.percentile(s, 1) + 1e-6)
+    s = np.clip(s, 0, 1)
+    return (s * 255).astype(np.uint8)
+
+# =========================
+# 输入区域（Upload or Sample）
+# =========================
+c1, c2, c3 = st.columns([2, 2, 1])
+
+if src == "Upload":
+    with c1:
+        vol_file = st.file_uploader(vol_label, type=["nii", "nii.gz", "mha"], key="vol")
+    with c2:
+        mask_file = st.file_uploader("Mask volume (.nii/.nii.gz/.mha)", type=["nii", "nii.gz", "mha"], key="mask")
+    with c3:
+        region = st.radio("Region", ["Head and Neck", "Abdomen", "Thorax"], index=1)
+    inputs_ready = (vol_file is not None) and (mask_file is not None)
+    region_for_run = region
+else:
+    with c1:
+        sample_key = st.selectbox("Choose a sample", list(SAMPLE_MAP.keys()))
+    with c2:
+        st.markdown("Region (fixed by sample)")
+        st.write(f"**{SAMPLE_MAP[sample_key]['region']}**")
+    with c3:
+        st.markdown(" ", unsafe_allow_html=True)
+    inputs_ready = (sample_key is not None)
+    region_for_run = SAMPLE_MAP[sample_key]["region"]
+
+run_btn = st.button("Run", type="primary", disabled=not inputs_ready)
+
+# =========================
+# 推理
+# =========================
+if run_btn:
+    with st.spinner(f"Running nnUNetv2 {('SynthradAlgorithm1' if task=='Task 1 (MR → CT)' else 'SynthradAlgorithm2')}..."):
+        if src == "Upload":
+            in_vol_img = _read_sitk_from_uploaded(vol_file)
+            mask_img = _read_sitk_from_uploaded(mask_file)
+        else:
+            sample = SAMPLE_MAP[sample_key]
+            in_vol_img = _read_sitk_from_path(sample["vol"])
+            mask_img = _read_sitk_from_path(sample["mask"])
+
+        # 保存原始元信息
+        st.session_state.orig_meta = (
+            in_vol_img.GetSpacing(),
+            in_vol_img.GetOrigin(),
+            in_vol_img.GetDirection(),
+        )
+
+        # 调用不同任务的算法
+        # 约定：算法统一使用 dict 输入：{"image": <sitk.Image>, "mask": <sitk.Image>, "region": <str>}
+        out_img = algo.predict({"image": in_vol_img, "mask": mask_img, "region": region_for_run})
+
+        st.session_state.synth_ct = out_img
+        st.session_state.vol_np = sitk.GetArrayFromImage(out_img).astype(np.float32)
+        st.session_state.input_vol = in_vol_img
+        st.session_state.input_mask = mask_img
+
+# =========================
+# 结果与下载
+# =========================
+if st.session_state.vol_np is None:
+    st.info("请先选择任务与输入（Upload 或 Sample），然后点击 Run")
+else:
+    # 将输出转为 LPS 方向做显示（可选）
+    out_lps = sitk.DICOMOrient(st.session_state.synth_ct, "LPS")
+    vol = sitk.GetArrayFromImage(out_lps).astype(np.float32)
+    D, H, W = vol.shape
+
+    col_d1, col_d2, col_d3 = st.columns(3)
+
+    # 下载合成CT
+    with col_d3:
+        _download_sitk_image(
+            st.session_state.synth_ct,
+            file_name="synth_ct.nii.gz",
+            label="Download synthetic CT",
+        )
+
+    # 下载输入体积（根据任务区分命名）
+    with col_d1:
+        if st.session_state.input_vol is not None:
+            in_name = "input_mri.nii.gz" if task == "Task 1 (MR → CT)" else "input_cbct.nii.gz"
+            in_label = "Download input MRI" if task == "Task 1 (MR → CT)" else "Download input CBCT"
+            _download_sitk_image(
+                st.session_state.input_vol,
+                file_name=in_name,
+                label=in_label,
+            )
+        else:
+            st.button("Download input", disabled=True)
+
+    # 下载掩��
+    with col_d2:
+        if st.session_state.input_mask is not None:
+            _download_sitk_image(
+                st.session_state.input_mask,
+                file_name="input_mask.nii.gz",
+                label="Download input Mask",
+            )
+        else:
+            st.button("Download input Mask", disabled=True)