SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1606.03475 · 2016

De-identification of Patient Notes with Recurrent Neural Networks

arXiv · PDF · Open in the Atlas

Code that ran

7 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 7 have no contract verdict; 0 failures of our instrument; 4 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/1606.03475.json · badge: /api/ran/1606.03475.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
Franck-Dernoncourt/NeuroNER community (classifier label)
classifier basis not recorded on the link
7 ran · 4 no recorded run
FunctionStatusWhere it lives
get_stanford_annotations Ran
ran on a synthesized input
Franck-Dernoncourt/NeuroNER/neuroner/brat_to_conll.py
code served (permissive licence) · get_code("b5925067ab578297")
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 json
from unittest.mock import Mock

# Create a mock CoreNLP object that simulates the Stanford CoreNLP annotator
class MockCoreNLP:
    def annotate(self, text, properties=None):
        # Simulate a typical CoreNLP JSON output
        output = {
            "sentences": [
                {
                    "index": 0,
                    "tokens": [
                        {
                            "index": 1,
                            "word": "The",
                            "originalText": "The",
                            "lemma": "the",
                            "characterOffsetBegin": 0,
                            "characterOffsetEnd": 3,
                            "pos": "DT"
                        },
                        {
                            "index": 2,
                            "word": "cat",
                            "originalText": "cat",
                            "lemma": "cat",
                            "characterOffsetBegin": 4,
                            "characterOffsetEnd": 7,
                            "pos": "NN"
                        }
                    ]
                }
            ]
        }
        return json.dumps(output)

# Create instances for the function arguments
text = "The cat sat on the mat."
core_nlp = MockCoreNLP()
port = 9000
annotators = 'tokenize,ssplit,pos,lemma'

ARGS = {
    'text': text,
    'core_nlp': core_nlp,
    'port': port,
    'annotators': annotators
}
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
is_token_in_pretrained_embeddings Ran
ran on a synthesized input
Franck-Dernoncourt/NeuroNER/neuroner/utils_nlp.py
code served (permissive licence) · get_code("4ba9c43926a08fce")
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 bool, 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 re

# Construct arguments for is_token_in_pretrained_embeddings
token = "hello"
all_pretrained_tokens = {"hello", "world", "test", "HELLO", "h3llo"}
parameters = {
    'check_for_lowercase': True,
    'check_for_digits_replaced_with_zeros': True
}

ARGS = {
    'token': token,
    'all_pretrained_tokens': all_pretrained_tokens,
    'parameters': parameters
}
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 bool, 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
load_pretrained_token_embeddings Ran
ran on a synthesized input
Franck-Dernoncourt/NeuroNER/neuroner/utils_nlp.py
code served (permissive licence) · get_code("224defd09e8e5ea5")
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 codecs
import numpy as np
import tempfile
import os

# Create a temporary embedding file
temp_dir = tempfile.mkdtemp()
embedding_file = os.path.join(temp_dir, "embeddings.txt")

# Write a small pretrained embedding file
with codecs.open(embedding_file, 'w', 'UTF-8') as f:
    f.write("hello 0.1 0.2 0.3 0.4\n")
    f.write("world 0.5 0.6 0.7 0.8\n")
    f.write("test 0.9 1.0 1.1 1.2\n")

ARGS = {
    'parameters': {
        'token_pretrained_embedding_filepath': embedding_file
    }
}
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 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
load_tokens_from_pretrained_token_embeddings Ran
ran on a synthesized input
Franck-Dernoncourt/NeuroNER/neuroner/utils_nlp.py
code served (permissive licence) · get_code("0069795317e6d20f")
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 set. 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 codecs
import tempfile
import os

# Create a temporary embedding file
temp_dir = tempfile.mkdtemp()
embedding_filepath = os.path.join(temp_dir, "embeddings.txt")

# Write a small pretrained embedding file
with codecs.open(embedding_filepath, 'w', 'UTF-8') as f:
    f.write("the 0.1 0.2 0.3 0.4\n")
    f.write("cat 0.5 0.6 0.7 0.8\n")
    f.write("dog 0.9 1.0 1.1 1.2\n")
    f.write("bird 1.3 1.4 1.5 1.6\n")

ARGS = {
    'parameters': {
        'token_pretrained_embedding_filepath': embedding_filepath
    }
}
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 set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
order_dictionary Ran
ran on a synthesized input
Franck-Dernoncourt/NeuroNER/neuroner/utils.py
code served (permissive licence) · get_code("73fc763bfb1aa699")
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 OrderedDict. 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 collections
import operator

# Create a sample dictionary with various key-value pairs
sample_dict = {
    'zebra': 3,
    'apple': 1,
    'mango': 2,
    'banana': 4
}

# Define ARGS mapping all parameters of order_dictionary
ARGS = {
    'dictionary': sample_dict,
    'mode': 'value',
    'reverse': False
}
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
parse_arguments Ran
ran on a synthesized input
Franck-Dernoncourt/NeuroNER/neuroner/__main__.py
code served (permissive licence) · get_code("848f9b6b6e76e0bd")
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 argparse
from argparse import RawTextHelpFormatter
import sys

# Since parse_arguments takes an optional 'arguments' parameter that defaults to None,
# we need to construct what would be passed to it. When arguments=None, argparse uses sys.argv[1:].
# We'll construct a list of command-line style arguments.

ARGS = {
    'arguments': [
        '--parameters_filepath', '/path/to/params.txt',
        '--character_embedding_dimension', '25',
        '--character_lstm_hidden_state_dimension', '25',
        '--check_for_digits_replaced_with_zeros', 'True',
        '--check_for_lowercase', 'True',
        '--dataset_text_folder', '/path/to/dataset',
        '--debug', 'False',
        '--dropout_rate', '0.5',
        '--experiment_name', 'test_experiment',
        '--freeze_token_embeddings', 'False',
        '--gradient_clipping_value', '5.0',
        '--learning_rate', '0.001',
        '--load_only_pretrained_token_embeddings', 'False',
        '--load_all_pretrained_token_embeddings', 'False',
        '--main_evaluation_mode', 'conll',
        '--maximum_number_of_epochs', '100',
        '--number_of_cpu_threads', '4',
        '--number_of_gpus', '0',
        '--optimizer', 'adam',
        '--output_folder', '/path/to/output',
        '--output_scores', 'True',
        '--patience', '10',
        '--plot_format', 'png',
        '--pretrained_model_folder', '/path/to/pretrained',
        '--reload_character_embeddings', 'False',
        '--reload_character_lstm', 'False',
        '--reload_crf', 'False',
        '--reload_feedforward', 'False',
        '--reload_token_embeddings', 'False',
        '--reload_token_lstm', 'False',
        '--remap_unknown_tokens_to_unk', 'True',
        '--spacylanguage', 'en',
        '--tagging_format', 'iob',
        '--token_embedding_dimension', '100',
        '--token_lstm_hidden_state_dimension', '100',
        '--token_pretrained_embedding_filepath', '/path/to/embeddings.txt',
        '--tokenizer', 'spacy',
        '--train_model', 'True',
        '--use_character_lstm', 'True',
        '--use_crf', 'True',
        '--use_pretrained_model', 'False',
        '--verbose', 'True',
        '--fetch_data', '',
        '--fetch_trained_model', '',
    ]
}
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
reverse_dictionary Ran
ran on a synthesized input
Franck-Dernoncourt/NeuroNER/neuroner/utils.py
code served (permissive licence) · get_code("9c479c5c207423da")
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 collections

# Test with a regular dictionary
regular_dict = {'a': 1, 'b': 2, 'c': 3}

# Test with an OrderedDict
ordered_dict = collections.OrderedDict([('x', 10), ('y', 20), ('z', 30)])

# We'll use the regular dict for this example, but both are valid
ARGS = {
    'dictionary': regular_dict
}
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
bidirectional_LSTM Not yet run
unverified (harvested, no recorded run)
Franck-Dernoncourt/NeuroNER/neuroner/entity_lstm.py
code served (permissive licence) · get_code("c13340f71e587c5b")
How it ran

no invocation recorded.

get_sentences_and_tokens_from_spacy Not yet run
unverified (harvested, no recorded run)
Franck-Dernoncourt/NeuroNER/neuroner/brat_to_conll.py
code served (permissive licence) · get_code("571b21d5e0204c30")
How it ran

no invocation recorded.

get_start_and_end_offset_of_token_from_spacy Not yet run
unverified (harvested, no recorded run)
Franck-Dernoncourt/NeuroNER/neuroner/brat_to_conll.py
code served (permissive licence) · get_code("0398487181c48d59")
How it ran

no invocation recorded.

renamed_load Not yet run
unverified (harvested, no recorded run)
Franck-Dernoncourt/NeuroNER/neuroner/utils.py
code served (permissive licence) · get_code("d16e6ce39db7140e")
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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 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.

FunctionAlso present inEarliest-datedOthers named
renamed_load 2 other papers
2 later
2003.01332 (2020-03)
Heterogeneous Graph Transformer
2006.15437 (2020-06)
GPT-GNN: Generative Pre-Training of Graph Neural 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

Objective: Patient notes in electronic health records (EHRs) may contain critical information for medical investigations. However, the vast majority of medical investigators can only access de-identified notes, in order to protect the confidentiality of patients. In the United States, the Health Insurance Portability and Accountability Act (HIPAA) defines 18 types of protected health information (PHI) that needs to be removed to de-identify patient notes. Manual de-identification is impractical given the size of EHR databases, the limited number of researchers with access to the non-de-identified notes, and the frequent mistakes of human annotators. A reliable automated de-identification system would consequently be of high value. Materials and Methods: We introduce the first de-identification system based on artificial neural networks (ANNs), which requires no handcrafted features or rules, unlike existing systems. We compare the performance of the system with state-of-the-art systems on two datasets: the i2b2 2014 de-identification challenge dataset, which is the largest publicly available de-identification dataset, and the MIMIC de-identification dataset, which we assembled and is twice as large as the i2b2 2014 dataset. Results: Our ANN model outperforms the state-of-the-art systems. It yields an F1-score of 97.85 on the i2b2 2014 dataset, with a recall 97.38 and a precision of 97.32, and an F1-score of 99.23 on the MIMIC de-identification dataset, with a recall 99.25 and a precision of 99.06. Conclusion: Our findings support the use of ANNs for de-identification of patient notes, as they show better performance than previously published systems while requiring no feature engineering.

For agents

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

get_harvested_code_for_paper("1606.03475")
get_code_for_paper("1606.03475")
have("1606.03475")

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

curl https://syntology.ai/api/ran/1606.03475.json

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

[![Syntology run record](https://syntology.ai/api/ran/1606.03475.svg)](https://syntology.ai/paper/1606.03475)

Connect an agent — have() is free.