SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.06956 · 2022

DermX: an end-to-end framework for explainable automated dermatological diagnosis

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; 0 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 0 failures of our instrument; 3 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/2202.06956.json · badge: /api/ran/2202.06956.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
ralucaj/dermx canonical (classifier label)
classifier basis not recorded on the link
7 ran · 3 no recorded run
FunctionStatusWhere it lives
create_and_save_class_map Ran
ran on a synthesized input
ralucaj/dermx/experiments/utils.py
code served (permissive licence) · get_code("40a295ac25760b0f")
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 os
import json
import tempfile

# Create a temporary directory for saving files
temp_dir = tempfile.mkdtemp()

# Define arguments for create_and_save_class_map
classes = ['cat', 'dog', 'bird', 'fish']
class_map_name = 'animals'
save_dir = temp_dir

ARGS = {
    'classes': classes,
    'class_map_name': class_map_name,
    'save_dir': save_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 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
create_characteristics_df Ran
ran on a synthesized input
ralucaj/dermx/experiments/analysis_utils.py
code served (permissive licence) · get_code("6f89bddc997c4964")
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 DataFrame. 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 pandas as pd

# Construct the evals dictionary with the expected structure
evals = {
    'labeller1_image1': {
        'annotations': [
            {'name': 'Acral distribution'},
            {'name': 'Closed comedo'},
            {'name': 'Cyst'},
        ]
    },
    'labeller1_image2': {
        'annotations': [
            {'name': 'Dark skin'},
            {'name': 'Female'},
            {'name': 'Macule'},
        ]
    },
    'labeller2_image1': {
        'annotations': [
            {'name': 'Hyperkeratosis'},
            {'name': 'Scale'},
        ]
    },
    'labeller2_image3': {
        'annotations': [
            {'name': 'Papule'},
            {'name': 'Male'},
            {'name': 'Young'},
        ]
    },
}

ARGS = {
    'evals': evals,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
create_diagnosis_df Ran
ran on a synthesized input
ralucaj/dermx/experiments/analysis_utils.py
code served (permissive licence) · get_code("4a81142fa07f61fc")
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 DataFrame. 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 pandas as pd

# Create sample evaluation data
evals = {
    'labeller1_img001': {
        'annotations': [
            {'name': 'Acne'},
            {'name': 'some_other_field'}
        ]
    },
    'labeller1_img002': {
        'annotations': [
            {'name': 'Psoriasis'}
        ]
    },
    'labeller2_img001': {
        'annotations': [
            {'name': 'Viral warts'}
        ]
    },
    'labeller2_img003': {
        'annotations': [
            {'name': 'unrelated_annotation'}
        ]
    },
    'labeller3_img002': {
        'annotations': [
            {'name': 'Vitiligo'}
        ]
    }
}

ARGS = {
    'evals': evals
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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_denormalize_transform Ran
ran on a synthesized input
ralucaj/dermx/experiments/datasets.py
code served (permissive licence) · get_code("f612470fbdaee4b0")
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 Normalize. 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
from torchvision import transforms as T

torch.manual_seed(0)

ARGS = {
    "mean_to_denorm": [0.485, 0.456, 0.406],
    "std_to_denorm": [0.229, 0.224, 0.225],
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Normalize — a Normalize output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Normalize 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
iou_metric Ran
ran on a synthesized input
ralucaj/dermx/experiments/train_utils.py
code served (permissive licence) · get_code("346b7bacc32597eb")
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 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Create 4D boolean tensors (batch_size=2, channels=3, height=4, width=5)
output = torch.randint(0, 2, (2, 3, 4, 5)).bool()
target = torch.randint(0, 2, (2, 3, 4, 5)).bool()

ARGS = {
    "output": output,
    "target": target,
    "sum_dims": (-2, -1),
    "epsilon": 1e-07,
    "per_characteristic": False,
}
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 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
make_dir Ran
ran on a synthesized input
ralucaj/dermx/experiments/utils.py
code served (permissive licence) · get_code("cc91b8b0cc1a93ac")
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 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.

import os
import tempfile

# Create a temporary directory to use as the base directory
temp_base_dir = tempfile.mkdtemp()

ARGS = {
    "dir": temp_base_dir,
    "name": "test_subdir"
}
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 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
read_darwin_evaluations Ran
ran on a synthesized input
ralucaj/dermx/experiments/analysis_utils.py
code served (permissive licence) · get_code("4b3b6dd35b0bffcf")
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 os
import json
import tempfile
from pathlib import Path

# Create a temporary directory with sample JSON files
temp_dir = tempfile.mkdtemp()

# Create sample evaluation JSON files
sample_evaluations = {
    "image_001.json": {
        "image_name": "image_001.jpg",
        "annotations": [
            {"id": 1, "label": "cat", "confidence": 0.95},
            {"id": 2, "label": "dog", "confidence": 0.87}
        ]
    },
    "image_002.json": {
        "image_name": "image_002.jpg",
        "annotations": [
            {"id": 3, "label": "bird", "confidence": 0.92}
        ]
    },
    "image_003.json": {
        "image_name": "image_003.jpg",
        "annotations": []
    }
}

# Write sample JSON files to the temporary directory
for filename, data in sample_evaluations.items():
    filepath = Path(temp_dir) / filename
    with open(filepath, 'w') as f:
        json.dump(data, f)

ARGS = {
    "export_path": 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 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
get_git_info Not yet run
unverified (harvested, no recorded run)
ralucaj/dermx/experiments/utils.py
code served (permissive licence) · get_code("325d7f42d225dba5")
How it ran

no invocation recorded.

iterate_through_attn_dataloader Not yet run
unverified (harvested, no recorded run)
ralucaj/dermx/experiments/train_utils.py
code served (permissive licence) · get_code("7b187e79f74a729b")
How it ran

no invocation recorded.

iterate_through_dataloader Not yet run
unverified (harvested, no recorded run)
ralucaj/dermx/experiments/train_utils.py
code served (permissive licence) · get_code("26a1291b7bd83a29")
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

Dermatological diagnosis automation is essential in addressing the high prevalence of skin diseases and critical shortage of dermatologists. Despite approaching expert-level diagnosis performance, convolutional neural network (ConvNet) adoption in clinical practice is impeded by their limited explainability, and by subjective, expensive explainability validations. We introduce DermX and DermX+, an end-to-end framework for explainable automated dermatological diagnosis. DermX is a clinically-inspired explainable dermatological diagnosis ConvNet, trained using DermXDB, a 554 image dataset annotated by eight dermatologists with diagnoses, supporting explanations, and explanation attention maps. DermX+ extends DermX with guided attention training for explanation attention maps. Both methods achieve near-expert diagnosis performance, with DermX, DermX+, and dermatologist F1 scores of 0.79, 0.79, and 0.87, respectively. We assess the explanation performance in terms of identification and localization by comparing model-selected with dermatologist-selected explanations, and gradient-weighted class-activation maps with dermatologist explanation maps, respectively. DermX obtained an identification F1 score of 0.77, while DermX+ obtained 0.79. The localization F1 score is 0.39 for DermX and 0.35 for DermX+. These results show that explainability does not necessarily come at the expense of predictive power, as our high-performance models provide expert-inspired explanations for their diagnoses without lowering their diagnosis performance.

For agents

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

get_harvested_code_for_paper("2202.06956")
get_code_for_paper("2202.06956")
have("2202.06956")

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

curl https://syntology.ai/api/ran/2202.06956.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.06956.svg)](https://syntology.ai/paper/2202.06956)

Connect an agent — have() is free.