SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.00538 · 2026

Vesselpose: Vessel Graph Reconstruction from Learned Voxel-wise Direction Vectors in 3D Vascular Images

Lisa Mais, Peter Hirsch, Dagmar Kainmueller, Rajalakshmi Palaniappan, Christoph Karg, Nemesio Navarro-Arambula, Kristin Kräker, C Palaniappan, N Karg, P Navarro-Arambula, K Hirsch, L Kräker, and 2 more

arXiv · PDF · Open in the Atlas

Code that ran

7 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; 6 failures of our instrument; 3 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/2605.00538.json · badge: /api/ran/2605.00538.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
Kainmueller-Lab/Vesselpose community
Papers with Code archive
7 ran · 3 no recorded run
FunctionStatusWhere it lives
build_kd_tree Ran
ran; our contract draft was wrong, not the code
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("99d7657afd2a86f1")
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) 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 numpy as np
from scipy.spatial import KDTree

# Construct edges_parent_radius argument
# The function expects an iterable of tuples: (edge, parent, radius)
# where edge is a tuple of two points (each point is array-like)

edges_parent_radius = [
    (
        (np.array([0.0, 1.0, 2.0, 3.0]), np.array([1.0, 2.0, 3.0, 4.0])),
        "parent_0",
        0.5
    ),
    (
        (np.array([2.0, 3.0, 4.0, 5.0]), np.array([3.0, 4.0, 5.0, 6.0])),
        "parent_1",
        0.7
    ),
    (
        (np.array([4.0, 5.0, 6.0, 7.0]), np.array([5.0, 6.0, 7.0, 8.0])),
        "parent_2",
        0.9
    ),
]

ARGS = {
    "edges_parent_radius": edges_parent_radius
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy — 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
calculate_distance Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("30d9b0148e5ab698")
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) 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 numpy as np

# Construct valid arguments for calculate_distance function
# The function takes a node (point) and an edge (pair of points)
# and finds the closest point on the edge to the node

# Define a node as a 2D point
node = [3.0, 4.0]

# Define an edge as a pair of 2D points (start and end)
edge = ([0.0, 0.0], [6.0, 0.0])

ARGS = {
    "node": node,
    "edge": edge
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — 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
find_reference_point Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("ce3bd10192574d7c")
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) 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

torch.manual_seed(0)

# Construct arguments for find_reference_point
# The function expects:
# - start_node: a point (list or array-like)
# - closest_point: a point (list or array-like)
# - step_size: a scalar float
# - parent: a tuple of two points (p1, p2)
# - edge: a tuple representing an edge (for comparison with parent)

start_node = [1.0, 2.0, 3.0, 4.0]
closest_point = [0.5, 1.5, 2.5, 3.5]
step_size = 2.5
parent = ([1.0, 2.0, 3.0, 4.0], [0.0, 0.0, 0.0, 0.0])
edge = ([1.0, 2.0, 3.0, 4.0], [0.0, 0.0, 0.0, 0.0])

ARGS = {
    "start_node": start_node,
    "closest_point": closest_point,
    "step_size": step_size,
    "parent": parent,
    "edge": edge,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
find_valid_closest_edge Ran
ran; our contract draft was wrong, not the code
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("232cfa6450568082")
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 scipy.spatial import cKDTree

torch.manual_seed(0)

# Create sample nodes and edges
node = [1.0, 2.0, 3.0, 4.0]

# Create edges: each edge is a tuple of two points (start, end)
edge1 = ([0.0, 1.0, 2.0, 3.0], [2.0, 3.0, 4.0, 5.0])
edge2 = ([1.5, 2.5, 3.5, 4.5], [3.0, 4.0, 5.0, 6.0])

# Create parent edges
parent1 = ([0.0, 0.0, 0.0, 0.0], [1.0, 1.0, 1.0, 1.0])
parent2 = ([1.0, 1.0, 1.0, 1.0], [2.0, 2.0, 2.0, 2.0])

# Create edges_parent_radius list
# Each entry is (edge, parent, radius)
edges_parent_radius = [
    (edge1, parent1, 10.0),
    (edge2, parent2, 15.0),
]

# Build KDTree from edge start points
edge_points = np.array([np.array(e[0]) for e in [edge1, edge2]])
kd_tree = cKDTree(edge_points)

# Parameters for the function
search_radius = 5.0
step_size = 2.0
radius_divisor = 2.0

ARGS = {
    'node': node,
    'edges_parent_radius': edges_parent_radius,
    'kd_tree': kd_tree,
    'search_radius': search_radius,
    'step_size': step_size,
    'radius_divisor': radius_divisor,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, 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
get_nifti_stem Ran
ran; our contract draft was wrong, not the code
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("df0538b215b85f65")
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) and returned str. 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.

from pathlib import Path

ARGS = {
    'path': Path('sample.nii.gz')
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
get_num_processes Ran
ran, honoured the contract we drafted
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("65479012151913b7")
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) 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.

from multiprocessing import Pool, cpu_count

ARGS = {
    'cfg': {
        'processing': {
            'num_processes': 'auto'
        }
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
load_config Ran
ran; our contract draft was wrong, not the code
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("57a1b166aaa595be")
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) 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 tempfile
import os
import yaml

# Create a temporary YAML config file
temp_dir = tempfile.mkdtemp()
config_path = os.path.join(temp_dir, 'config.yaml')

config_data = {
    'dataset_type': 'mnist',
    'processing_by_dataset': {
        'mnist': {
            'normalize': True,
            'augment': False,
            'batch_size': 32
        },
        'cifar10': {
            'normalize': True,
            'augment': True,
            'batch_size': 64
        }
    }
}

with open(config_path, 'w') as f:
    yaml.dump(config_data, f)

ARGS = {
    'config_path': config_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
yaml — 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
clear_folder_contents Not yet run
unverified (harvested, no recorded run)
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("e35f58549f7fad27")
How it ran

no invocation recorded.

find_shortest_distance_to_edge Not yet run
unverified (harvested, no recorded run)
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("1a86ff38dc713b62")
How it ran

no invocation recorded.

main Not yet run
unverified (harvested, no recorded run)
Kainmueller-Lab/Vesselpose/vector/vector_generation.py
code served (permissive licence) · get_code("14d2110d81843986")
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

Blood vessel segmentation and -tracing are essential tasks in many medical imaging applications. Although numerous methods exist, the prevailing segment-then-fix paradigm is fundamentally limited regarding its suitability for modeling the task of complete and topologically accurate vascular network reconstruction. Here, we propose an approach to extract topologically more accurate vascular graphs from 3D image data, building upon highly successful ideas from the related biomedical tasks of cell segmentation and -tracking. Our approach first predicts voxel-wise vessel direction vectors joint with standard vessel segmentation masks. Second, to extract the vascular graph from these predictions, we introduce a direction-vector-guided extension of the TEASAR algorithm. Our approach achieves state-of-the-art performance on three benchmark datasets, spanning both synthetic and real imagery. We further demonstrate the applicability of our approach to challenging 3D micro-CT scans of rat heart vasculature. Finally, we propose meaningful and interpretable measures of topological error, namely false splits and false merges for graphs. Overall, our approach substantially improves the topological accuracy of reconstructed vascular graphs, being able to separate closely apposed vessel segments and handle multiple vascular trees within a single volume.

For agents

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

get_harvested_code_for_paper("2605.00538")
get_code_for_paper("2605.00538")
have("2605.00538")

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

curl https://syntology.ai/api/ran/2605.00538.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.00538.svg)](https://syntology.ai/paper/2605.00538)

Connect an agent — have() is free.