Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,49 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
---
|
| 4 |
+
# PROSPERO Inclusion/Exclusion Criteria Dataset
|
| 5 |
+
|
| 6 |
+
This dataset is a curated and preprocessed collection of clinical research objectives and their corresponding inclusion and exclusion criteria, extracted from the [PROSPERO](https://www.crd.york.ac.uk/prospero/) international prospective register of systematic reviews.
|
| 7 |
+
|
| 8 |
+
## Description
|
| 9 |
+
|
| 10 |
+
The dataset was constructed to support fine-tuning of large language models (LLMs) for tasks involving the automated generation of eligibility criteria based on research objectives. It contains English-language research objectives from systematic review protocols across three domains:
|
| 11 |
+
|
| 12 |
+
- Mental health
|
| 13 |
+
- Diagnostics
|
| 14 |
+
- Therapeutics
|
| 15 |
+
|
| 16 |
+
## Dataset Structure
|
| 17 |
+
|
| 18 |
+
Each row in the dataset corresponds to one study protocol and contains:
|
| 19 |
+
|
| 20 |
+
- `Study_ID`: The unique identifier from PROSPERO (e.g., CRD420250616527)
|
| 21 |
+
- `Objectives`: The research objective(s) of the systematic review
|
| 22 |
+
- `Included`: The criteria defining which participants/studies are eligible for inclusion
|
| 23 |
+
- `Excluded`: The criteria used to disqualify participants/studies from inclusion
|
| 24 |
+
|
| 25 |
+
|
| 26 |
+
## Use Case
|
| 27 |
+
|
| 28 |
+
This dataset is ideal for training or fine-tuning LLMs (like T5, Mistral, LLaMA) on:
|
| 29 |
+
|
| 30 |
+
- Generating inclusion/exclusion criteria from research objectives
|
| 31 |
+
- Biomedical text-to-text reasoning tasks
|
| 32 |
+
- Automating systematic review assistance and meta-research workflows
|
| 33 |
+
|
| 34 |
+
## Format
|
| 35 |
+
|
| 36 |
+
- File type: CSV (`utf-8`)
|
| 37 |
+
- Fields: `Study_ID`, `Objectives`, `Included`, `Excluded`
|
| 38 |
+
- Suggested fields for fine-tuning:
|
| 39 |
+
- `input_text`: Prompt generated from `Objectives`
|
| 40 |
+
- `output_text`: "Inclusion: ... Exclusion: ..." combined string
|
| 41 |
+
|
| 42 |
+
## License & Attribution
|
| 43 |
+
|
| 44 |
+
Data is sourced from PROSPERO, a publicly accessible database, and is intended for educational and research purposes only. Please cite the original PROSPERO records when using this dataset in any published work.
|
| 45 |
+
|
| 46 |
+
## 🙌 Created By
|
| 47 |
+
|
| 48 |
+
Sanaa as part of her mission to automate and enhance systematic literature reviews using AI.
|
| 49 |
+
email: sanaa.abril@gmail.com
|