SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.05735 · 2021

The Medical Segmentation Decathlon

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2106.05735.json · badge: /api/ran/2106.05735.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow community (classifier label)
classifier basis not recorded on the link
9 ran · 0 no recorded run
FunctionStatusWhere it lives
dice_coefficient Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/segmentation_losses.py
code served (permissive licence) · get_code("55c491331fb03c16")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned SymbolicTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import tensorflow.keras.backend as K
import numpy as np

# Set seed for reproducibility
np.random.seed(0)
tf.random.set_seed(0)

# Create small tensors with shape (batch_size=2, x_dim=4, y_dim=3, z_dim=2, num_classes=5)
batch_size = 2
x_dim = 4
y_dim = 3
z_dim = 2
num_classes = 5

# Create ground truth tensor with random binary values
y_true_np = np.random.randint(0, 2, size=(batch_size, x_dim, y_dim, z_dim, num_classes)).astype(np.float32)
y_true = K.variable(y_true_np)

# Create predictions tensor with random values between 0 and 1
y_pred_np = np.random.uniform(0, 1, size=(batch_size, x_dim, y_dim, z_dim, num_classes)).astype(np.float32)
y_pred = K.variable(y_pred_np)

# Define axis for reduction
axis = (0, 1, 2, 3)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
    "axis": axis,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
dice_coefficient_binary Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/segmentation_losses.py
code served (permissive licence) · get_code("429dba7efc52d89d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned SymbolicTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Set seeds for reproducibility
np.random.seed(0)
tf.random.set_seed(0)

# Create small binary segmentation tensors
# Shape: (batch_size=2, x_dim=4, y_dim=3, z_dim=5)
batch_size = 2
x_dim = 4
y_dim = 3
z_dim = 5

# Ground truth: binary segmentation masks
y_true = tf.constant(
    np.random.randint(0, 2, size=(batch_size, x_dim, y_dim, z_dim)),
    dtype=tf.float32
)

# Predictions: continuous values between 0 and 1
y_pred = tf.constant(
    np.random.rand(batch_size, x_dim, y_dim, z_dim),
    dtype=tf.float32
)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
dice_loss Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/segmentation_losses.py
code served (permissive licence) · get_code("a036a5593e08b1b0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned SymbolicTensor. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import tensorflow.keras.backend as K
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create small tensors matching the expected shape
# (batch_size=2, x_dim=4, y_dim=3, z_dim=2, num_classes=5)
batch_size = 2
x_dim = 4
y_dim = 3
z_dim = 2
num_classes = 5

# Create ground truth tensor with random values between 0 and 1
y_true = tf.constant(
    np.random.rand(batch_size, x_dim, y_dim, z_dim, num_classes).astype(np.float32)
)

# Create prediction tensor with random values between 0 and 1
y_pred = tf.constant(
    np.random.rand(batch_size, x_dim, y_dim, z_dim, num_classes).astype(np.float32)
)

# Define the axis parameter
axis = (0, 1, 2, 3)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
    "axis": axis,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
generate_model_prediction Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/generate_prediction_files_for_challenge_submission.py
code served (permissive licence) · get_code("3712ead23ff0f9b1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family int, shape (4, 5, 6) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a minimal mock model that has a predict method
class MockSegmentationModel(nn.Module):
    def __init__(self, binary_mode=True):
        super().__init__()
        self.binary_mode = binary_mode
        self.fc = nn.Linear(1, 1)
    
    def predict(self, x):
        # x shape: (batch, dim_x, dim_y, dim_z, num_channels)
        batch_size = x.shape[0]
        dim_x = x.shape[1]
        dim_y = x.shape[2]
        dim_z = x.shape[3]
        
        if self.binary_mode:
            # For binary mode, output shape should be (batch, dim_x, dim_y, dim_z, 1)
            return np.random.rand(batch_size, dim_x, dim_y, dim_z, 1).astype(np.float32)
        else:
            # For multi-class mode, output shape should be (batch, dim_x, dim_y, dim_z, num_classes)
            num_classes = 3
            return np.random.rand(batch_size, dim_x, dim_y, dim_z, num_classes).astype(np.float32)

# Create test MRI data with shape (dim_x, dim_y, dim_z, num_channels)
mri_data = np.random.randn(4, 5, 6, 2).astype(np.float32)

# Create models for both binary and non-binary modes
model_binary = MockSegmentationModel(binary_mode=True)
model_multiclass = MockSegmentationModel(binary_mode=False)

# Define arguments for the function
ARGS = {
    "model": model_binary,
    "mri": mri_data,
    "binary_mode": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (4, 5, 6) — shape (4, 5, 6) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
set_tensorflow_mirrored_strategy_gpu_devices_list Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/train_unet_segmentation_model.py
code served (permissive licence) · get_code("ba859058bfab9035")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

ARGS = {
    'num_gpus': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
standardize Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/generate_prediction_files_for_challenge_submission.py
code served (permissive licence) · get_code("4e50db17fbb87fc5")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 4, dtype family float, shape (4, 3, 2, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small MRI array with shape (dim_x, dim_y, dim_z, num_channels)
# Using distinct sizes: 4, 3, 2, 5 for each dimension
mri = np.random.randn(4, 3, 2, 5).astype(np.float32)

ARGS = {
    "mri": mri
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 4, dtype family float, shape (4, 3, 2, 5) — shape (4, 3, 2, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
unet_3d_conv3dtranspose_dropout Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/model.py
code served (permissive licence) · get_code("85683318706e43ef")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Functional. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Input, Conv3D, MaxPooling3D, UpSampling3D, Dropout, Conv3DTranspose, BatchNormalization, concatenate

ARGS = {
    "input_size": (240, 240, 160, 4),
    "unet_resize_factor": 2,
    "unet_dropout_rate": 0.3,
    "num_classes": 4,
    "binary_model": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
unet_3d_upsampling_batchnormalization Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/model.py
code served (permissive licence) · get_code("0364327269701c67")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Functional. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Input, Conv3D, MaxPooling3D, UpSampling3D, Dropout, Conv3DTranspose, BatchNormalization, concatenate

ARGS = {
    "input_size": (240, 240, 160, 4),
    "unet_resize_factor": 2,
    "num_classes": 4,
    "binary_model": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
unet_3d_upsampling_dropout Ran
ran on a synthesized input
tamerthamoqa/3D-mri-brain-tumour-image-segmentation-medical-decathlon-tensorflow/model.py
code served (permissive licence) · get_code("533b4851f6eba632")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Functional. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Input, Conv3D, MaxPooling3D, UpSampling3D, Dropout, Conv3DTranspose, BatchNormalization, concatenate

# Set seeds for reproducibility
tf.random.set_seed(0)

ARGS = {
    "input_size": (240, 240, 160, 4),
    "unet_resize_factor": 2,
    "unet_dropout_rate": 0.3,
    "num_classes": 4,
    "binary_model": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

International challenges have become the de facto standard for comparative assessment of image analysis algorithms given a specific task. Segmentation is so far the most widely investigated medical image processing task, but the various segmentation challenges have typically been organized in isolation, such that algorithm development was driven by the need to tackle a single specific clinical problem. We hypothesized that a method capable of performing well on multiple tasks will generalize well to a previously unseen task and potentially outperform a custom-designed solution. To investigate the hypothesis, we organized the Medical Segmentation Decathlon (MSD) - a biomedical image analysis challenge, in which algorithms compete in a multitude of both tasks and modalities. The underlying data set was designed to explore the axis of difficulties typically encountered when dealing with medical images, such as small data sets, unbalanced labels, multi-site data and small objects. The MSD challenge confirmed that algorithms with a consistent good performance on a set of tasks preserved their good average performance on a different set of previously unseen tasks. Moreover, by monitoring the MSD winner for two years, we found that this algorithm continued generalizing well to a wide range of other clinical problems, further confirming our hypothesis. Three main conclusions can be drawn from this study: (1) state-of-the-art image segmentation algorithms are mature, accurate, and generalize well when retrained on unseen tasks; (2) consistent algorithmic performance across multiple tasks is a strong surrogate of algorithmic generalizability; (3) the training of accurate AI segmentation models is now commoditized to non AI experts.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2106.05735")
get_code_for_paper("2106.05735")
have("2106.05735")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2106.05735.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2106.05735.svg)](https://syntology.ai/paper/2106.05735)

Connect an agent — have() is free.