Khyatimirani/womens-health-fertility-safety-qa
Dataset Card for WomenHealth-Ferility-safety-guideline This dataset card aims to be a base template for new datasets. It has been generated using this raw template. Dataset Details Dataset Description This dataset contains clinically grounded question–answer pairs focused on fertility treatments and IVF (In Vitro Fertilization), with a strong emphasis on patient safety, procedure awareness, and emotional reassurance. The dataset is designed to… See the full description on the dataset page: https://huggingface.co/datasets/Khyatimirani/womens-health-fertility-safety-qa.
Dataset Card for WomenHealth-Ferility-safety-guideline
<!-- Provide a quick summary of the dataset. -->
This dataset card aims to be a base template for new datasets. It has been generated using this raw template.
Dataset Details
Dataset Description
This dataset contains clinically grounded question–answer pairs focused on fertility treatments and IVF (In Vitro Fertilization), with a strong emphasis on patient safety, procedure awareness, and emotional reassurance.
The dataset is designed to address high-frequency, anxiety-driven questions asked by IVF patients across different stages of treatment, including:
Pre-IVF evaluation and stimulation
Hormonal injections and side effects
Egg retrieval (OPU) and recovery
Embryo transfer (ET)
OHSS awareness and risk monitoring
Lifestyle, movement, diet, and daily activity safety during treatment
Emotional health, uncertainty, and waiting periods
All responses are non-diagnostic, safety-first, and structured to support AI systems that must operate responsibly in sensitive clinical contexts.
- Curated by: Khyati Mirani
- Funded by [optional]: Not Funded
- Shared by [optional]: Independent open-source contribution
- Language(s) (NLP): English
- License: CC BY 4.0 (Creative Commons Attribution 4.0)
Dataset Sources [optional]
<!-- Provide the basic links for the dataset. -->
- Repository: Hugging Face Datasets Hub
- Paper [optional]: (https://www.eshre.eu/)
Uses
<!-- Address questions around how the dataset is intended to be used. -->
Direct Use
This dataset is intended for:
IVF clinic chatbots and AI fertility companions
Safety-focused patient education tools in reproductive medicine
Reducing repetitive non-emergency questions handled by clinicians
Fine-tuning LLMs for clinical escalation vs reassurance decisions
Training systems to distinguish:
Normal vs concerning symptoms
Self-care vs “contact your doctor immediately” cases
Research on patient anxiety patterns during fertility treatments
⚠️ The dataset is not a substitute for medical care and must be deployed with clinician oversight.
[More Information Needed]
Out-of-Scope Use
This dataset should not be used for:
Medical diagnosis or treatment planning
IVF protocol personalization without clinician approval
Emergency medical decision-making
Insurance, legal, or outcome prediction systems
Fertility success rate estimation or embryo grading
[More Information Needed]
Dataset Structure
Each row in the dataset represents a single IVF-related clinical safety Q&A. | Field Name | Type | Description | | --------------------------- | ------------- | ------------------------------------------------------------------------------------------- | | id | string | Unique identifier for the Q&A entry | | question | string | Primary patient-facing question | | semantic_variants | list[string] | Alternative phrasings of the same question reflecting patient anxiety or wording variations | | answer | string | Safety-first, empathetic, non-diagnostic response | | ivf_stage | enum | null | IVF journey stage associated with the question | | category | enum | High-level classification of the question | | requires_doctor_attention | boolean | Indicates whether the patient should contact a doctor | | escalation_reason | string | null | Reason for escalation when doctor attention is required | | symptoms_mentioned | list[string] | Symptoms explicitly referenced in the question | | keywords | list[string] | Medical, procedural, or safety-related keywords | | source | string | Origin of the Q&A (e.g., curated, clinical education, guideline-inspired) |
ivfstage Enum | Value | Description | | ----------------------- | -------------------------------------------------- | | `beforestimulation | Before hormonal stimulation begins | | duringstimulation` | During ovarian stimulation injections | | `triggerdaypreopu | Trigger day and pre–egg retrieval window | | duringopu` | Egg retrieval (OPU) procedure day | | `afteropu | Recovery period following egg retrieval | | betweenopuandet` | Interval between egg retrieval and embryo transfer | | `afterembryo_transfer | Post–embryo transfer (two-week wait period) | | general` | Not specific to a single IVF stage |
category Enum | Value | Description | | -------------- | ----------------------------------------------- | | general_faq | Common informational fertility or IVF questions | | medication | Hormonal injections, medicines, side effects | | procedure | IVF-related procedures (OPU, ET, scans, tests) | | lifestyle | Movement, diet, travel, daily activity guidance | | warning_sign | Symptoms that may require medical attention | | emergency | High-risk symptoms requiring immediate care |
⚠️ Safety Rule (Hard Constraint)
If category is warning_sign or emergency:
requiresdoctorattention must be true
escalation_reason must be populated
This rule enables:
Safe AI behavior
Clear escalation logic
Separation of reassurance vs clinical risk
Dataset Creation
Curation Rationale
Curation Rationale
IVF patients often experience:
High emotional stress
Information overload
Uncertainty about what symptoms are normal vs dangerous
Repetitive questioning during sensitive clinical windows
This dataset was created to:
Provide consistent, doctor-aligned safety messaging
Reduce unnecessary panic while not suppressing red flags
Enable AI systems to escalate correctly when needed
Support humane, empathetic communication in fertility care
Source Data
Questions were derived from:
Common IVF patient FAQs
Clinical workflow pain points
Observed message overload during procedures
Answers were crafted to:
Avoid clinical decision-making
Emphasize monitoring, reassurance, and escalation criteria
Use patient-friendly, calming language
Data was manually curated, validated, and structured.
Data Collection and Processing
Dataset authored by an independent health-tech researcher and engineer
Content informed by clinical IVF workflows and patient education practices
No real patient chats or records were used
Who are the source data producers?
<!-- This section describes the people or systems who originally created the data. It should also include self-reported demographic or identity information for the source data creators if this information is available. -->
[More Information Needed]
Annotations [optional]
Annotation process
Annotation process
Manual labeling of:
IVF stages
Risk vs non-risk scenarios
Lifestyle and movement safety
Semantic variants added to capture:
Anxious phrasing
Reassurance-seeking language
Keywords included to support retrieval and escalation logic
Who are the annotators?
Single-curator dataset (author-generated annotations)
Personal and Sensitive Information
Personal and Sensitive Information
No personal or identifiable patient data is included
No real medical records or transcripts were used
All content is synthetic, generalized, and educational
Bias, Risks, and Limitations
IVF protocols vary by clinic, country, and patient profile
Dataset reflects general safety guidance, not protocol-specific rules
Emotional framing may not cover all cultural contexts
English-only in the current version
Recommendations
Always pair deployment with:
Clear medical disclaimers
Doctor-approved response layers
Emergency escalation pathways
Best used as:
A first-line educational and reassurance system
Not as a decision-maker
Clinics should customize messaging to their protocols
Citation [optional]
@dataset{womenshealthivfclinicalsafety_qa, title={womens-health-ivf-clinical-safety-qa}, author={Mirani, Khyati}, year={2026}, publisher={Hugging Face} }
BibTeX:
[More Information Needed]
APA:
[More Information Needed]
Glossary [optional]
Mirani, K. (2026). womens-health-ivf-clinical-safety-qa. Hugging Face Datasets.
More Information [optional]
Glossary [optional]
IVF: In Vitro Fertilization
OPU: Ovum Pick-Up (Egg Retrieval)
ET: Embryo Transfer
OHSS: Ovarian Hyperstimulation Syndrome
Clinical Safety Q&A: Responses designed to reassure while identifying red flags
Dataset Card Authors [optional]
[More Information Needed]
Dataset Card Contact
[More Information Needed]
