SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.05201 · 2020

ROSE: A Retinal OCT-Angiography Vessel Segmentation Dataset and New Model

arXiv · PDF · Open in the Atlas

Code that ran

10 of 11 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 8 have no contract verdict; 2 failures of our instrument; 1 no recorded run), measured 2026-11-21; 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/2007.05201.json · badge: /api/ran/2007.05201.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
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network community (classifier label)
classifier basis not recorded on the link
10 ran · 1 no recorded run
FunctionStatusWhere it lives
Crop Ran
ran on a synthesized input
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/fundus_dataset.py
code served (permissive licence) · get_code("bca69db90e38dc43")
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 tuple. 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 torch
import numpy as np
from PIL import Image
from torchvision import transforms
import torchvision.transforms.functional as TF

torch.manual_seed(0)

# Create small PIL images for image, label, and mask
# The function expects PIL Images based on usage with TF.crop
# Create images with size larger than crop_size so cropping is possible
image_array = np.random.randint(0, 256, (128, 128, 3), dtype=np.uint8)
image = Image.fromarray(image_array)

label_array = np.random.randint(0, 256, (128, 128, 3), dtype=np.uint8)
label = Image.fromarray(label_array)

# mask needs to have areas with value 255 for the function to find valid crops
# Create a mask where most pixels are 255 (valid region)
mask_array = np.full((128, 128), 255, dtype=np.uint8)
mask = Image.fromarray(mask_array)

ARGS = {
    "image": image,
    "label": label,
    "mask": mask,
    "max_attempts": 3,
    "crop_size": (64, 64)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple 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
build_loss Ran
ran on a synthesized input
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/losses.py
code served (permissive licence) · get_code("70bab76a9adb7f32")
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 focal_loss. 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 torch
import torch.nn as nn

torch.manual_seed(0)

# The function build_loss takes a single parameter 'loss' which is a string
# that determines which loss criterion to instantiate.
# Valid options are: 'mse', 'l1', 'smoothl1', 'bce', 'focal', 'dice'

ARGS = {
    'loss': 'focal'
}
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 focal_loss — a focal_loss output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a focal_loss 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
calc_auc Ran
ran on a synthesized input
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/evaluation.py
code served (permissive licence) · get_code("f1a1dc397cac1ca3")
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 float, ndim 0, dtype family float, shape () 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
from sklearn import metrics

np.random.seed(0)

# Create sample prediction array (probabilities/scores)
pred_arr = np.array([
    [0.1, 0.9, 0.3, 0.7],
    [0.2, 0.8, 0.4, 0.6]
])

# Create sample ground truth array (binary labels)
gt_arr = np.array([
    [0, 1, 0, 1],
    [0, 1, 1, 0]
])

# Create sample mask array (0 means exclude, non-zero means include)
mask_arr = np.array([
    [1, 1, 0, 1],
    [1, 0, 1, 1]
])

ARGS = {
    "pred_arr": pred_arr,
    "gt_arr": gt_arr,
    "mask_arr": mask_arr
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () 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
clip_by_tensor Ran
ran on a synthesized input
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/losses.py
code served (permissive licence) · get_code("a6980e3ba0fcef1c")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Create sample tensors
t = torch.randn(2, 4)  # Small tensor to clip
t_min = torch.tensor(-1.0)  # Minimum threshold
t_max = torch.tensor(1.0)   # Maximum threshold

ARGS = {
    "t": t,
    "t_min": t_min,
    "t_max": t_max
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
deconv Ran
ran; our contract draft was wrong, not the code
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/imed_models.py
code served (permissive licence) · get_code("b528298ef3ced2c0")
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 ConvTranspose2d. 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 torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_channels": 3,
    "out_channels": 64,
}
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 ConvTranspose2d — a ConvTranspose2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ConvTranspose2d 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
extract_mask Ran
ran on a synthesized input
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/evaluation.py
code served (permissive licence) · get_code("32bab0b40aa74387")
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 tuple, ndim 2, 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 numpy as np

# Create small numpy arrays for testing
pred_arr = np.array([[0.1, 0.9, 0.3, 0.7],
                      [0.2, 0.8, 0.4, 0.6]])

gt_arr = np.array([[0, 1, 0, 1],
                    [0, 1, 1, 0]])

mask_arr = np.array([[1, 1, 0, 1],
                      [1, 0, 1, 1]])

ARGS = {
    'pred_arr': pred_arr,
    'gt_arr': gt_arr,
    'mask_arr': mask_arr
}
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 tuple, ndim 2, dtype family float — a tuple 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
linear_stretch Ran
ran on a synthesized input
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/fundus_dataset.py
code served (permissive licence) · get_code("f65d403bcfc1c4b1")
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 2, dtype family float, shape (4, 3) 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 test image array with distinct values
# Using a 4x3 array to have unambiguous dimensions
img_arr = np.array([
    [10.0, 20.0, 30.0],
    [40.0, 50.0, 60.0],
    [70.0, 80.0, 90.0],
    [100.0, 110.0, 120.0]
], dtype=np.float32)

ARGS = {
    "img_arr": img_arr
}
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 2, dtype family float, shape (4, 3) — shape (4, 3) 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
paint_border_overlap Ran
ran on a synthesized input
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/fundus_dataset.py
code served (permissive licence) · get_code("d44fedcfc25f202f")
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 2, dtype family float, shape (104, 124) 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 2D grayscale image
grayscale_img = np.random.rand(100, 120).astype(np.float32)

# Create a 3D RGB image
rgb_img = np.random.rand(100, 120, 3).astype(np.float32)

ARGS = {
    "full_img": grayscale_img,
    "crop_size": (64, 64),
    "stride": (5, 5),
}
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 2, dtype family float, shape (104, 124) — shape (104, 124) 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
random_crop Ran
ran on a synthesized input
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/octa_dataset.py
code served (permissive licence) · get_code("42834384f9d9373b")
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 tuple. 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 torch
import torchvision.transforms.functional as TF
from PIL import Image
import numpy as np

torch.manual_seed(0)

# Create a small PIL Image for the image parameter
# RandomCrop.get_params expects a PIL Image with size attribute
image_array = np.random.randint(0, 256, (512, 512, 3), dtype=np.uint8)
image = Image.fromarray(image_array)

# Create a label image (same size as image)
label_array = np.random.randint(0, 256, (512, 512), dtype=np.uint8)
label = Image.fromarray(label_array, mode='L')

ARGS = {
    'image': image,
    'label': label,
    'crop_size': (256, 256)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple 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
short_hash Ran
ran; our contract draft was wrong, not the code
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/resnest.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from RobertHong1992/Resnest, is pointer only (licence: NONE) · get_code("979123ace229d553")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from RobertHong1992/Resnest, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

numeric_score Not yet run
unverified (harvested, no recorded run)
iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network/code/OCTA-Net/evaluation.py
code served (permissive licence) · get_code("3abdf9144e7b2061")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 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.

FunctionAlso present inEarliest-datedOthers named
deconv 1 other paper
1 later
2306.05671 (2023-06)
Topology-Aware Uncertainty for Image Segmentation
short_hash 1 other paper
1 dated earlier
2004.08955 (2020-04)
ResNeSt: Split-Attention Networks

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Optical Coherence Tomography Angiography (OCT-A) is a non-invasive imaging technique, and has been increasingly used to image the retinal vasculature at capillary level resolution. However, automated segmentation of retinal vessels in OCT-A has been under-studied due to various challenges such as low capillary visibility and high vessel complexity, despite its significance in understanding many eye-related diseases. In addition, there is no publicly available OCT-A dataset with manually graded vessels for training and validation. To address these issues, for the first time in the field of retinal image analysis we construct a dedicated Retinal OCT-A SEgmentation dataset (ROSE), which consists of 229 OCT-A images with vessel annotations at either centerline-level or pixel level. This dataset has been released for public access to assist researchers in the community in undertaking research in related topics. Secondly, we propose a novel Split-based Coarse-to-Fine vessel segmentation network (SCF-Net), with the ability to detect thick and thin vessels separately. In the SCF-Net, a split-based coarse segmentation (SCS) module is first introduced to produce a preliminary confidence map of vessels, and a split-based refinement (SRN) module is then used to optimize the shape/contour of the retinal microvasculature. Thirdly, we perform a thorough evaluation of the state-of-the-art vessel segmentation models and our SCF-Net on the proposed ROSE dataset. The experimental results demonstrate that our SCF-Net yields better vessel segmentation performance in OCT-A than both traditional methods and other deep learning methods.

For agents

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

get_harvested_code_for_paper("2007.05201")
get_code_for_paper("2007.05201")
have("2007.05201")

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

curl https://syntology.ai/api/ran/2007.05201.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.05201.svg)](https://syntology.ai/paper/2007.05201)

Connect an agent — have() is free.