ID
stringlengths 32
32
| URL
stringlengths 92
152
| Product_Name
stringlengths 4
43
| Full_Content
stringlengths 27.3k
1.43M
| Section_1
stringlengths 107
48.9k
| Section_2
stringlengths 93
84.2k
| Section_3
stringlengths 88
67.6k
| Section_4
stringlengths 66
131k
| Section_5
stringlengths 87
32.7k
| Section_6
stringlengths 66
103k
|
---|---|---|---|---|---|---|---|---|---|
5C706E5A9E28B681F969CE8D6DC55607 | https://www.ema.europa.eu/documents/product-information/pergoveris-epar-product-information_en.pdf | Pergoveris | "1\n\nANNEX I\n\nSUMMARY OF PRODUCT CHARACTERISTICS\n\n\n\n2\n\n1. NAME OF THE MEDICINAL PRODUCT\n\n(...TRUNCATED) | "{'Title': '1. what pergoveris is and what it is used for', 'Section_Content': 'what pergoveris is p(...TRUNCATED) | "{'Title': '2. what you need to know before you use pergoveris', 'Section_Content': 'you and your pa(...TRUNCATED) | "{'Title': '3. how to use pergoveris', 'Section_Content': 'always use this medicine exactly as your (...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': 'like all medicines, this medicine can cau(...TRUNCATED) | "{'Title': '5. how to store pergoveris', 'Section_Content': 'keep this medicine out of the sight and(...TRUNCATED) | "{'Title': '6. contents of the pack and other information', 'Section_Content': \"what pergoveris con(...TRUNCATED) |
17F9F8809EB0FC0F70E195786F80368D | https://www.ema.europa.eu/documents/product-information/refludan-epar-product-information_en.pdf | Refludan | "1 \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n\n \n(...TRUNCATED) | "{'Title': '1. what refludan is and what it is used for', 'Section_Content': 'refludan is an antithr(...TRUNCATED) | "{'Title': '2. before you use refludan', 'Section_Content': 'do not use refludan - if you are allerg(...TRUNCATED) | "{'Title': '3. how to use refludan', 'Section_Content': 'your doctor will determine and control the (...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': 'like all medicines, refludan can have sid(...TRUNCATED) | {'Title': '5 how to store refludan', 'Section_Content': None, 'Entity_Recognition': None} | "{'Title': '6. further information', 'Section_Content': 'what refludan contains the active substance(...TRUNCATED) |
4263665B0FB8C7F2A0039D03A3678097 | https://www.ema.europa.eu/documents/product-information/onsenal-epar-product-information_en.pdf | Onsenal | "1 \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\nANNEX I \n \n\nSUMMARY(...TRUNCATED) | "{'Title': '1. what onsenal is and what it is used for', 'Section_Content': 'onsenal belongs to a gr(...TRUNCATED) | "{'Title': '2. before you take onsenal', 'Section_Content': 'do not take onsenal - if you have had a(...TRUNCATED) | "{'Title': '3. how to take onsenal', 'Section_Content': 'always take onsenal exactly as your doctor (...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': 'like all medicines, onsenal can cause sid(...TRUNCATED) | "{'Title': '5. how to store onsenal', 'Section_Content': \"keep out of the reach and sight of childr(...TRUNCATED) | "{'Title': '6. further information', 'Section_Content': \"what onsenal contains − the active subst(...TRUNCATED) |
0B6B1FB1EF2BF8BE2D880507FBFFF23E | https://www.ema.europa.eu/documents/product-information/nerlynx-epar-product-information_en.pdf | Nerlynx | "1 \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\nANNEX I \n \n\nSUMMARY(...TRUNCATED) | "{'Title': '1. what nerlynx is and what it is used for', 'Section_Content': \"what nerlynx is nerlyn(...TRUNCATED) | "{'Title': '2. what you need to know before you take nerlynx', 'Section_Content': 'do not take nerly(...TRUNCATED) | "{'Title': '3. how to take nerlynx', 'Section_Content': 'always take this medicine exactly as your d(...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': 'like all medicines, this medicine can cau(...TRUNCATED) | "{'Title': '5. how to store nerlynx', 'Section_Content': 'keep this medicine out of the sight and re(...TRUNCATED) | "{'Title': '6. contents of the pack and other information', 'Section_Content': \"what nerlynx contai(...TRUNCATED) |
843352215D2AEF5460300234FD08912B | https://www.ema.europa.eu/documents/product-information/oncaspar-epar-product-information_en.pdf | Oncaspar | "1 \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\nANNEX I \n \n\nSUMMARY(...TRUNCATED) | "{'Title': '1. what oncaspar is and what it is used for', 'Section_Content': 'oncaspar contains pega(...TRUNCATED) | "{'Title': '2. what you need to know before you are given oncaspar', 'Section_Content': 'do not use (...TRUNCATED) | "{'Title': '3. how oncaspar is given', 'Section_Content': 'your treatment with oncaspar has been pre(...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': 'like all medicines, this medicine can cau(...TRUNCATED) | "{'Title': '5. how to store oncaspar', 'Section_Content': 'keep this medicine out of the sight and r(...TRUNCATED) | "{'Title': '6. contents of the pack and other information', 'Section_Content': 'what oncaspar contai(...TRUNCATED) |
1BC9051B5DB082D98245D790AE3F3E75 | "https://www.ema.europa.eu/documents/product-information/ivabradine-anpharm-epar-product-information(...TRUNCATED) | Ivabradine Anpharm | "1 \nClassified as confidential by the European Medicines Agency \n\n \n \n \n \n \n \n \n \n \n \n (...TRUNCATED) | "{'Title': '1. what ivabradine anpharm is and what it is used for', 'Section_Content': 'ivabradine a(...TRUNCATED) | "{'Title': '2. what you need to know before you take ivabradine anpharm', 'Section_Content': 'do not(...TRUNCATED) | "{'Title': '3. how to take ivabradine anpharm', 'Section_Content': 'always take this medicine exactl(...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': 'like all medicines, this medicine can cau(...TRUNCATED) | "{'Title': '5 how to store ivabradine anpharm', 'Section_Content': \"keep this medicine out of the s(...TRUNCATED) | "{'Title': '6. contents of the pack and other information', 'Section_Content': 'what ivabradine anph(...TRUNCATED) |
9A689AE7A7A56E0E3DE36E5A2B02ED5C | https://www.ema.europa.eu/documents/product-information/fasturtec-epar-product-information_en.pdf | Fasturtec | "ANNEX I \n\n \n\nSUMMARY OF PRODUCT CHARACTERISTICS \n\n \n\n\n\n2 \n\n1. NAME OF THE MEDICINAL PRO(...TRUNCATED) | "{'Title': '1. what fasturtec is and what it is used for', 'Section_Content': 'fasturtec contains th(...TRUNCATED) | "{'Title': '2. what you need to know before you are given fasturtec', 'Section_Content': 'do not use(...TRUNCATED) | "{'Title': '3. how to use fasturtec', 'Section_Content': 'fasturtec is to be given to you before or (...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': 'like all medicines, this medicine can cau(...TRUNCATED) | "{'Title': '5. how to store fasturtec', 'Section_Content': 'keep this medicine out of the sight and (...TRUNCATED) | "{'Title': '6. contents of the pack and other information', 'Section_Content': 'what fasturtec conta(...TRUNCATED) |
59BBDD1548644AF445C0E04FF1A79925 | "https://www.ema.europa.eu/documents/product-information/pemetrexed-sandoz-epar-product-information_(...TRUNCATED) | Pemetrexed Sandoz | "1 \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\nANNEX I \n \n\nSUMMARY(...TRUNCATED) | "{'Title': '1. what pemetrexed sandoz is and what it is used for', 'Section_Content': 'pemetrexed sa(...TRUNCATED) | "{'Title': '2. what you need to know before you use pemetrexed sandoz', 'Section_Content': 'do not u(...TRUNCATED) | "{'Title': '3. how to use pemetrexed sandoz', 'Section_Content': \"the dose of pemetrexed sandoz is (...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': 'like all medicines, this medicine can cau(...TRUNCATED) | "{'Title': '5. how to store pemetrexed sandoz', 'Section_Content': 'keep this medicine out of the si(...TRUNCATED) | "{'Title': '6. contents of the pack and other information', 'Section_Content': 'what pemetrexed sand(...TRUNCATED) |
21983D0EEABCE47D7E24D8FE76233F0D | https://www.ema.europa.eu/documents/product-information/liprolog-epar-product-information_en.pdf | Liprolog | "1 \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\nANNEX I \n \n\nSUMMARY(...TRUNCATED) | "{'Title': '1. what liprolog is and what it is used for', 'Section_Content': 'liprolog is used to tr(...TRUNCATED) | "{'Title': '2. what you need to know before you use liprolog', 'Section_Content': 'do not use liprol(...TRUNCATED) | "{'Title': '3. how to use liprolog', 'Section_Content': 'always check the pack and the vial label fo(...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': \"like all medicines, this medicine can ca(...TRUNCATED) | "{'Title': '5. how to store liprolog', 'Section_Content': 'before the first use store your liprolog (...TRUNCATED) | "{'Title': '6. contents of the pack and other information', 'Section_Content': \"what liprolog 100 u(...TRUNCATED) |
F1F6F4F82FDF85EBBFD5EB5CBCBA820C | "https://www.ema.europa.eu/documents/product-information/kogenate-bayer-epar-product-information_en.(...TRUNCATED) | Kogenate Bayer | "1 \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\nANNEX I \n \n\nSUMMARY(...TRUNCATED) | "{'Title': '1. what kogenate bayer is and what it is used for', 'Section_Content': \"kogenate bayer (...TRUNCATED) | "{'Title': '2. what you need to know before you use kogenate bayer', 'Section_Content': 'do not use (...TRUNCATED) | "{'Title': '3. how to use kogenate bayer', 'Section_Content': 'always use this medicine exactly as d(...TRUNCATED) | "{'Title': '4. possible side effects', 'Section_Content': \"like all medicines, this medicine can ca(...TRUNCATED) | "{'Title': '5. how to store kogenate bayer', 'Section_Content': 'keep this medicine out of the sight(...TRUNCATED) | "{'Title': '6. contents of the pack and other information', 'Section_Content': 'what kogenate bayer (...TRUNCATED) |
Dataset Card for BioLeaflets Dataset
Dataset Summary
BioLeaflets is a biomedical dataset for Data2Text generation. It is a corpus of 1,336 package leaflets of medicines authorised in Europe, which were obtained by scraping the European Medicines Agency (EMA) website. Package leaflets are included in the packaging of medicinal products and contain information to help patients use the product safely and appropriately. This dataset comprises the large majority (∼ 90%) of medicinal products authorised through the centralised procedure in Europe as of January 2021. For more detailed information, please read the paper at ACL Anthology.
Supported Tasks and Leaderboards
BioLeaflets proposes a conditional generation task (data-to-text) in the biomedical domain: given an ordered set of entities as source, the goal is to produce a multi-sentence section. Successful generation thus requires the model to learn specific syntax, terminology, and writing style from the corpus. Alternatively, the dataset might be used for named-entity recognition task: given text, detect medical entities. The dataset provides an extensive description of medicinal products and thus supports a plain language modeling task focused on biomedical data.
Languages
Monolingual - en.
Dataset Structure
Data Instances
For each instance (leaflet), there is a unique ID, URL, Product_Name, and textual information clearly describing the medicine. The content of each section is not standardized (NO template), yet it is still well-structured. Each document contains six sections:
- What is the product and what is it used for
- What you need to know before you take the product
- Product usage instructions
- possible side effects
- product storage conditions
- other information
Every section is represented as a dictionary with the 'Title', 'Section_Content', and 'Entity_Recognition' as keys. The content of each section is lower-cased and tokenized by treating all special characters as separate tokens.
Data Fields
ID
: a string representing a unique ID assigned to a leafletURL
: a string containing the link to the article on the European Medicines Agency (EMA) websiteProduct Name
: a string, the name of the medicineFull Content
: a string covering the full content of the article available at URLSection 1
: a dictionary including section 1 content and associated medical entitiesSection 2
: a dictionary including section 2 content and associated medical entitiesSection 3
: a dictionary including section 3 content and associated medical entitiesSection 4
: a dictionary including section 4 content and associated medical entitiesSection 5
: a dictionary including section 5 content and associated medical entitiesSection 6
: a dictionary including section 6 content and associated medical entities
Data Splits
We randomly split the dataset into training (80%), development (10%), and test (10%) set. Duplicates are removed.
Dataset Creation
Curation Rationale
Introduce a new biomedical dataset (BioLeaflets), which could serve as a benchmark for biomedical text generation models. BioLeaflets proposes a conditional generation task: given an ordered set of entities as source, the goal is to produce a multi-sentence section.
Source Data
Initial Data Collection and Normalization
The dataset was obtained by scraping the European Medicines Agency (EMA) website. Each leaflet has an URL associated with it to the article on the EMA website.
Who are the source language producers?
Labeling experts with domain knowledge produced factual information.
Annotations
Annotation process
To create the required input for data-to-text generation, we augment each document by leveraging named entity recognition (NER). We combine two NER frameworks: Amazon Comprehend Medical (commercial) and Stanford Stanza (open-sourced). Additionally, we treat all digits as entities and add the medicine name as the first entity
Who are the annotators?
Machine-generated: ensemble of state-of-the-art named entity recognition (NER) models.
Personal and Sensitive Information
[Not included / Not present]
Considerations for Using the Data
Social Impact of Dataset
The purpose of this dataset is to help develop models that can automatically generate long paragraphs of text as well as to facilitate the development of NLP models in the biomedical domain. The main challenges of this dataset for D2T generation are multi-sentence and multi-section target text, small sample size, specialized medical vocabulary, and syntax.
Discussion of Biases
Package leaflets are published for medicinal products approved in the European Union (EU). They are included in the packaging of medicinal products and contain information to help patients use the product safely and appropriately. The dataset represents factual information produced by labeling experts and validated before publishing. Hence, biases of any kind are not present in the dataset. Package leaflets are required to be written in a way that is clear and understandable.
Other Known Limitations
[N/A]
Additional Information
Dataset Curators
The data was originally collected by Ruslan Yermakov*, Nicholas Drago, and Angelo Ziletti at Bayer AG (Decision Science & Advanced Analytics unit). The code is made publicly available at github link
* Work done during internship.
Licensing Information
The BioLeaflets dataset is released under the Apache-2.0 License.
Citation Information
@inproceedings{yermakov-etal-2021-biomedical,
title = "Biomedical Data-to-Text Generation via Fine-Tuning Transformers",
author = "Yermakov, Ruslan and
Drago, Nicholas and
Ziletti, Angelo",
booktitle = "Proceedings of the 14th International Conference on Natural Language Generation",
month = aug,
year = "2021",
address = "Aberdeen, Scotland, UK",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.inlg-1.40",
pages = "364--370",
abstract = "Data-to-text (D2T) generation in the biomedical domain is a promising - yet mostly unexplored - field of research. Here, we apply neural models for D2T generation to a real-world dataset consisting of package leaflets of European medicines. We show that fine-tuned transformers are able to generate realistic, multi-sentence text from data in the biomedical domain, yet have important limitations. We also release a new dataset (BioLeaflets) for benchmarking D2T generation models in the biomedical domain.",
}
Contributions
Thanks to @wingedRuslan for adding this dataset.
- Downloads last month
- 65