SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.20104 · 2026

NucFuseRank: Dataset Fusion and Performance Ranking for Nuclei Instance Segmentation

Nima Torbati, Anastasia Meshcheryakova, Ramona Woitek, Sepideh Hatamikia, Diana Mechtcheriakova, Amirreza Mahbod

arXiv · PDF · Open in the Atlas

Code that ran

6 of 10 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 4 no recorded run), measured 2026-11-22; 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/2601.20104.json · badge: /api/ran/2601.20104.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
masih4/NucleiAnalysis canonical (classifier label)
classifier basis: census
6 ran · 4 no recorded run
FunctionStatusWhere it lives
adapt_checkpoint Ran
ran; our contract draft was wrong, not the code
masih4/NucleiAnalysis/utils/utils.py
code served (permissive licence) · get_code("584f108a5ca12126")
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 dict. 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)

# Create a small model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

# Create a checkpoint dict with some matching and some mismatched keys
checkpoint = {
    'linear1.weight': torch.randn(3, 4),
    'linear1.bias': torch.randn(3),
    'linear2.weight': torch.randn(2, 3),
    'linear2.bias': torch.randn(2),
    'extra_key': torch.randn(5, 5),  # This will be skipped due to shape mismatch
}

ARGS = {
    'checkpoint': checkpoint,
    'model': model,
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
add_pad Ran
ran; our contract draft was wrong, not the code
masih4/NucleiAnalysis/Evaluate.py
code served (permissive licence) · get_code("76298b1a1e974f19")
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 torch.nn.functional as F

torch.manual_seed(0)

# Create a small input tensor for testing
# Shape: (batch=2, channels=3, height=4, width=5)
x = torch.randn(2, 3, 4, 5)

# Define pad_siz as a tuple (height_pad, width_pad)
pad_siz = (1, 2)

ARGS = {
    'x': x,
    'pad_siz': pad_siz
}
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 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
count_num_tiles Ran
ran, honoured the contract we drafted
masih4/NucleiAnalysis/datasets_stats.py
code served (permissive licence) · get_code("fb2af1b19ae4f01b")
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 int, ndim 0, dtype family int, 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 os
import tempfile
import shutil

# Create a temporary directory structure with fold0 and fold1
temp_dir = tempfile.mkdtemp()

# Create fold0/images directory with .tif files
fold0_dir = os.path.join(temp_dir, 'fold0', 'images')
os.makedirs(fold0_dir, exist_ok=True)
for i in range(3):
    open(os.path.join(fold0_dir, f'image_{i}.tif'), 'w').close()

# Create fold1/images directory with .tif files
fold1_dir = os.path.join(temp_dir, 'fold1', 'images')
os.makedirs(fold1_dir, exist_ok=True)
for i in range(2):
    open(os.path.join(fold1_dir, f'image_{i}.tif'), 'w').close()

# Also add some non-.tif files to verify filtering works
open(os.path.join(fold0_dir, 'readme.txt'), 'w').close()
open(os.path.join(fold1_dir, 'metadata.json'), 'w').close()

ARGS = {
    'pth': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, 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
fill_background_holes_batch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
masih4/NucleiAnalysis/utils/utils.py
code served (permissive licence) · get_code("9573e6692db0ea67")
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 (2, 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
import cv2

# Create a small batch of labeled masks
# Shape: (N, H, W) with values [0-5]
# N=2 (batch size), H=4, W=3 (height and width)
batch_size = 2
height = 4
width = 3

masks = np.array([
    # First mask in batch
    [
        [0, 1, 1],
        [1, 0, 1],
        [2, 2, 2],
        [2, 0, 2]
    ],
    # Second mask in batch
    [
        [0, 3, 3],
        [3, 0, 3],
        [4, 4, 4],
        [4, 0, 4]
    ]
], dtype=np.uint8)

max_hole_size = 5000

ARGS = {
    "masks": masks,
    "max_hole_size": max_hole_size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — 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 (2, 4, 3) — shape (2, 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
main Ran
ran; our contract draft was wrong, not the code
masih4/NucleiAnalysis/datasets_stats.py
code served (permissive licence) · get_code("28a2dcba7477caee")
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. 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 numpy as np
import os
import tempfile
import shutil

# Create a temporary directory structure with test data
temp_dir = tempfile.mkdtemp()
dataset_dir = os.path.join(temp_dir, 'dataset')
test_dir = os.path.join(temp_dir, 'test')
os.makedirs(dataset_dir, exist_ok=True)
os.makedirs(test_dir, exist_ok=True)

# Create sample .tif and .npy files in dataset directory
for i in range(3):
    # Create dummy mask data with different sizes
    if i == 0:
        mask = np.random.randint(0, 5, size=(128, 128))
    elif i == 1:
        mask = np.random.randint(0, 8, size=(256, 256))
    else:
        mask = np.random.randint(0, 3, size=(64, 64))
    
    # Save as .npy
    np.save(os.path.join(dataset_dir, f'image_{i}.npy'), mask)
    # Create corresponding .tif placeholder (just save as dummy)
    np.save(os.path.join(dataset_dir, f'image_{i}.tif.npy'), mask)
    open(os.path.join(dataset_dir, f'image_{i}.tif'), 'w').close()

# Create test files
for i in range(2):
    mask = np.random.randint(0, 4, size=(192, 192))
    np.save(os.path.join(test_dir, f'test_image_{i}.npy'), mask)
    open(os.path.join(test_dir, f'test_image_{i}.tif'), 'w').close()

class Args:
    def __init__(self, dataset_path, test_path, dataset_tag):
        self.dataset_path = dataset_path
        self.test_path = test_path
        self.dataset_tag = dataset_tag

args = Args(
    dataset_path=dataset_dir,
    test_path=test_dir,
    dataset_tag='test_image'
)

ARGS = {'args': args}
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 list — a list 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
remove_pad Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
masih4/NucleiAnalysis/EvaluateAll.py
code served (permissive licence) · get_code("5b94c1204c199768")
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 4, dtype family float, shape (2, 4, 7, 7) 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 numpy as np

torch.manual_seed(0)

# Create a padded tensor: (B=2, C=4, H_padded=10, W_padded=12)
x = torch.randn(2, 4, 10, 12)

# Define padding amounts: (pad_left, pad_right, pad_top, pad_bottom)
pad = (2, 3, 1, 2)

ARGS = {
    'x': x,
    'pad': pad
}
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 Tensor, ndim 4, dtype family float, shape (2, 4, 7, 7) — shape (2, 4, 7, 7) 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
adapt_checkpoint_dualEncoder Not yet run
unverified (harvested, no recorded run)
masih4/NucleiAnalysis/utils/utils.py
code served (permissive licence) · get_code("f8d862b1e4d06729")
How it ran

no invocation recorded.

calculate_instances Not yet run
unverified (harvested, no recorded run)
masih4/NucleiAnalysis/post_proc_cellvit.py
code served (permissive licence) · get_code("a54fa31e5cb698ae")
How it ran

no invocation recorded.

pad_to_multiple Not yet run
unverified (harvested, no recorded run)
masih4/NucleiAnalysis/Evaluate.py
code served (permissive licence) · get_code("002bda16c58651bc")
How it ran

no invocation recorded.

tissue_aug_gpu Not yet run
unverified (harvested, no recorded run)
masih4/NucleiAnalysis/utils/class_augs.py
code served (permissive licence) · get_code("ff96a9da222c0e73")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 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 10 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

Nuclei instance segmentation in hematoxylin and eosin (H&E)-stained images plays an important role in automated histological image analysis, with various applications in downstream tasks. While several machine learning and deep learning approaches have been proposed for nuclei instance segmentation, most research in this field focuses on developing new segmentation algorithms and benchmarking them on a limited number of arbitrarily selected public datasets. In this work, rather than focusing on model development, we focused on the datasets used for this task. Based on an extensive literature review, we

For agents

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

get_harvested_code_for_paper("2601.20104")
get_code_for_paper("2601.20104")
have("2601.20104")

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

curl https://syntology.ai/api/ran/2601.20104.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.20104.svg)](https://syntology.ai/paper/2601.20104)

Connect an agent — have() is free.