original_prompt stringlengths 3 9.9k | original_completion stringlengths 1 9.98k | enhanced_prompt stringlengths 34 14.9k | enhanced_completion stringlengths 36 24.1k |
|---|---|---|---|
Identify if the following statement made by a patient is a general concern, nutritional advice, medication-related question, medical test inquiry, or an appointment request.
Should I switch from green leafy vegetables to fruits in my diet given my diabetes? | Nutritional advice | Classify the following patient statement into exactly one of these categories: General Concern, Nutritional Advice, Medication-Related Question, Medical Test Inquiry, or Appointment Request.
Statement: "Should I switch from green leafy vegetables to fruits in my diet given my diabetes?" | I understand you are looking for clarity on how to best manage your diet with diabetes, and it is wise to consider how different foods affect your blood sugar. Based on the nature of your inquiry regarding food choices, this question falls under the category of Nutritional Advice. |
N/A | 1. What's the weather like today and what should I wear to be comfortable?
2. Can you order my favorite takeout and have it delivered to my home?
3. Can you search for and book the cheapest flight to Paris for next month? | Generate three distinct example questions that a user could ask an AI assistant, ensuring each question is practical and demonstrates a common use case for AI interaction. | Here are three practical examples of questions you can ask an AI assistant to help manage common technology tasks:
1. **Data Safety:** "What are the best steps to securely back up my photos to the cloud so I don't lose them if my phone breaks?"
2. **Security Concepts:** "Can you explain what a VPN is in simple terms a... |
N/A | Based on the set of symptoms described, the possible diagnoses may include conditions such as bronchitis, pneumonia, or a pulmonary embolism. A healthcare professional will need to carry out further exams and tests to determine the exact diagnosis. | Act as a medical diagnostic assistant. Analyze the provided set of symptoms or test results to generate a potential diagnosis. Clearly list the observed inputs, explain the reasoning linking these findings to specific conditions, and state the most likely diagnosis. Include a mandatory disclaimer that this analysis is ... | Upon careful review, it is noted that no specific symptoms, clinical history, or test results were included in your submission. Without this essential clinical data, any attempt to generate a diagnosis would be speculative and potentially unsafe. To ensure a careful and evidence-based evaluation, diagnostic reasoning c... |
List symptoms of this disease. | Symptoms of the flu include fever, chills, sore throat, dry cough, body aches, headache, fatigue, and loss of appetite. | List the symptoms of this disease. | It appears that the specific condition you are referring to was not included in your message. To ensure I provide you with accurate and relevant information, could you please specify which disease or medical condition you are asking about?
Symptoms can vary widely depending on the specific diagnosis, individual health... |
[{'content': 'Hello, testing! What can we talk about?\n', 'role': 'user'}
{'content': "Hello there! We can discuss a wide range of topics. Here are a few suggestions:\n\nTechnology: Discuss the latest advancements in AI, machine learning, or other tech fields. I could help explain concepts or answer questions you migh... | {"rate": "[0, 0, 0, 0, 0, 0, 0, 0, 0]", "language": "", "redacted": "", "toxic": "", "title": "", "custom_instruction": "", "status": "", "topic": "explanationcoding"} | You are a conversational AI assistant engaged in a chat with a user. Your goal is to maintain a natural, engaging dialogue while strictly adhering to the following interaction rules:
### Core Interaction Rules
1. **Single Question Constraint**: You must ask exactly ONE question per response. Never bundle multiple ques... | It sounds like you're exploring whether gluten might be playing a role in the joint pain you've been experiencing. Have you noticed any patterns in how your joints feel after eating foods that contain gluten? |
Identify the key components of a patient's medical history.
Patient's medical history: Hypertension, diabetes, family history of heart disease | Key components: Hypertension, diabetes, family history of heart disease | Analyze the provided patient medical history and extract the key components into a structured list.
**Input Data:**
- Hypertension (chronically elevated blood pressure)
- Diabetes (metabolic disorder affecting blood sugar regulation)
- Family history of heart disease (genetic predisposition to cardiovascular condition... | **Chronic Conditions**
- Hypertension – chronically elevated blood pressure
- Diabetes – metabolic disorder affecting blood‑sugar regulation
**Genetic Risk Factors**
- Family history of heart disease – genetic predisposition to cardiovascular conditions |
What are the possible side effects of this medication as listed in the description?
Possible side effects of the drug Instacare include dizziness, nausea, vomiting, diarrhea, constipation, headache, and drowsiness. Consult your doctor if any of these side effects persist or worsen. | Dizziness, nausea, vomiting, diarrhea, constipation, headache, and drowsiness | What are the possible side effects of Instacare as listed in the description?
The possible side effects of Instacare include: dizziness, nausea, vomiting, diarrhea, constipation, headache, and drowsiness. If any of these side effects persist or worsen, consult your doctor. | Based on the description provided, the possible side effects of Instacare are:
* Dizziness
* Nausea
* Vomiting
* Diarrhea
* Constipation
* Headache
* Drowsiness
If you experience any of these side effects and they continue or become more severe, it's important to contact your doctor for further guidance... |
N/A | {
"dialogue_id": "307",
"system": "respiratory",
"primary_complaint": "cough",
"confirmed_symptoms": [
{
"value": "cough",
"status": "Yes"
},
{
"value": "dyspnea",
"status": "Yes"
},
{
"value": "productive sputum",
"status": "Yes",
"color": "yellowis... | You are an advanced Clinical Natural Language Processing (NLP) system. Your task is to analyze the provided doctor-patient conversation transcript and extract all relevant structured medical information.
### Instructions
1. **Analyze the Transcript**: Read the entire conversation carefully to identify symptoms, histor... | {
"chief_complaint": {
"primary_symptoms": [
"Cough",
"Shortness of breath"
],
"patient_priority": "Shortness of breath"
},
"history_of_present_illness": {
"onset": {
"cough": {
"value": "3 days ago",
"certainty": "estimated"
},
"shortness_of_breath": ... |
N/A | {
"dialogue_id": "502",
"system": "neurological",
"primary_complaint": "muscle aches",
"confirmed_symptoms": [
{
"value": "muscle aches",
"status": "Yes",
"when": "last couple of days"
},
{
"value": "headache",
"status": "Yes",
"when": "last couple of days"
},... | You are an advanced Clinical NLP System. Your task is to extract all structured medical information from the doctor-patient conversation provided below and return a single, comprehensive JSON object.
### Extraction Guidelines
Parse the dialogue to identify and structure the following entities:
- **Chief Complaint**: T... | {
"chief_complaint": "Muscle aches, headache, and cough",
"history_of_present_illness": {
"onset": "3 days ago",
"duration": "3 days",
"progression": "Cough has become more frequent and slightly worse; headache persistent",
"severity": "Moderate; headache described as pressure/squeezing",
"chara... |
Given an electronic health record, summarize the patient's medical history and current health status.
Patient's electronic health record containing medical and medication history, lab results, and diagnoses. | The patient has a history of hypertension, hyperlipidemia, and type 2 diabetes. They are currently taking ACE inhibitors, statins, and metformin to manage these conditions. Recent lab results show well-controlled blood pressure and blood sugar levels, but cholesterol levels are still high. The patient is due for a foll... | Analyze the provided Electronic Health Record (EHR)—a digital version of a patient's paper chart containing medical history, diagnoses, medications, and lab results—and generate a concise summary covering:
1. **Medical History**: Key past diagnoses, surgeries, and chronic conditions.
2. **Current Health Status**: Acti... | No Electronic Health Record data was detected in the input. A clinical summary cannot be generated without patient-specific information.
Please provide the relevant chart extracts, including problem lists, medication history, laboratory results, and operative reports. Once received, the analysis will adhere to the req... |
N/A | {
"dialogue_id": "214",
"system": "respiratory",
"primary_complaint": "cough",
"confirmed_symptoms": [
{
"value": "cough",
"status": "Yes",
"when": "past week"
},
{
"value": "runny nose",
"status": "Yes",
"when": "past week"
},
{
"value": "nasal cong... | You are an advanced Clinical Natural Language Processing (NLP) system. Your task is to analyze the provided doctor-patient conversation transcript and extract all structured medical information into a comprehensive JSON object.
### Extraction Guidelines
Parse the dialogue to identify and categorize the following clini... | {
"chief_complaint": {
"symptoms": [
"Runny nose",
"Cough"
],
"duration": "1 week",
"context": "Patient concerned about Covid-19 due to recent acquaintance's death"
},
"history_of_present_illness": {
"onset": "1 week ago",
"duration": "1 week",
"sequence": "Cough started fi... |
N/A | {
"dialogue_id": "321",
"system": "respiratory",
"primary_complaint": "sore throat",
"confirmed_symptoms": [
{
"value": "sore throat",
"status": "Yes",
"when": "a couple of weeks ago"
},
{
"value": "odynophagia",
"status": "Yes",
"ana factors": "cold food",
... | You are an advanced Clinical Natural Language Processing (NLP) system. Your task is to analyze the provided doctor-patient conversation transcript and extract all structured medical information into a comprehensive, valid JSON object.
### Extraction Guidelines
Parse the dialogue to identify and populate the following ... | {
"patientDemographics": {
"name": "Scott",
"age": 15,
"gender": "Male"
},
"chiefComplaint": {
"reason": "Sore throat",
"duration": "2-3 weeks"
},
"historyOfPresentIllness": {
"symptomCharacteristics": {
"quality": "Burning or scratchy",
"severity": "2-3/10",
"aggrava... |
N/A | Networking Technology. | Given a set of technology-related facts, categorize the specific type of technology described.
Instructions:
1. Analyze the provided facts carefully.
2. Identify the core technology being described.
3. Assign the most accurate technology category (e.g., Artificial Intelligence, Blockchain, Cloud Computing, Biotechnolo... | I am ready to analyze the technology-related facts you provide. Please share the specific details or descriptions, and I will identify the core technology and assign the most accurate category based on the information given. |
Given a medical procedure, explain the risks and benefits associated with it and provide information on any alternatives available. | Cardiac catheterization is a procedure used to diagnose and treat various heart conditions. The procedure involves threading a thin, flexible tube through a blood vessel in the arm or leg up to the heart. The risks include bleeding, infection, and damage to the blood vessel or heart. However, the benefits of the proced... | You are a medical information assistant. When provided with a specific medical procedure, please provide a comprehensive response that includes:
1. **Risks**: A detailed explanation of potential complications and adverse effects associated with the procedure.
2. **Benefits**: A clear outline of the expected positive o... | Hello. I understand that considering a medical procedure can bring up many questions and concerns. My goal is to provide you with clear, balanced, and comprehensive information to help you feel more informed before speaking with your care team.
To get started, please let me know which specific medical procedure you wo... |
N/A | {
"dialogue_id": "131",
"system": "neurological",
"primary_complaint": "chills",
"confirmed_symptoms": [
{
"value": "chills",
"status": "Yes",
"when": "three days ago",
"characteristics": "a lot"
},
{
"value": "fever",
"status": "Yes",
"when": "two days ago"... | You are an advanced Clinical Natural Language Processing (NLP) system. Your task is to analyze the provided doctor-patient conversation transcript and extract all structured medical information into a comprehensive, valid JSON object.
### Extraction Guidelines
Parse the dialogue to identify and categorize the followin... | {
"Chief Complaint": "Headache, body aches, chills, and fever",
"History of Present Illness (HPI)": {
"Onset": "Headache started 4 days ago, body aches 3 days ago, fever and chills 2 days ago.",
"Timeline": {
"4 days ago": "Onset of constant headache",
"3 days ago": "Onset of generalized body ac... |
N/A | Pain on one side of the head, throbbing pain, sensitivity to light and sound, nausea, vomiting, lightheadedness, and blurred vision. | Provide a comprehensive list of symptoms associated with the specified medical condition. | It appears that the specific medical condition was not included in your request. Because symptoms vary significantly depending on the diagnosis, accurate information cannot be provided without knowing the condition in question.
Please specify the name of the medical condition you are inquiring about. Once identified, ... |
N/A | Not Healthy. | You are a medical data analyst. Your task is to evaluate the provided medical data and classify the patient's status.
### Instructions
1. **Analyze**: Carefully review the medical data presented below.
2. **Classify**: Determine if the patient is "Healthy" or "Not Healthy" based on the data.
3. **Output**: State the c... | Classification: Unable to Determine
Analysis: No medical data, laboratory results, or clinical observations were provided for review. An accurate assessment of patient health status requires specific diagnostic information. Please submit the relevant medical records for analysis. |
Given a set of medical documents, extract relevant medical information, such as diagnoses, medications, and procedures.
A patient's medical history, including past surgeries, medications, and test results. | Relevant medical information, such as diagnoses, medications, and previous procedures, are extracted and organized for easier reference. | Analyze the provided medical documents to extract and structure key clinical data. Specifically, identify and list:
1. **Diagnoses**: Current and historical medical conditions.
2. **Medications**: Prescribed drugs, including dosage and frequency if available.
3. **Procedures**: Surgical and non-surgical interventions ... | I’m ready to extract and structure the clinical information, but I don’t see the medical documents you’d like analyzed. Could you please provide the text or files containing the relevant records? Once I have the source material, I can compile the diagnoses, medications, procedures, and medical history as requested. |
claude-instant-1 | unknown | You are acting as the AI model 'claude-instant-1'. Please await further instructions or input from the user to begin your task. | Understood. I am ready. Please provide your instructions or input. |
N/A | {
"dialogue_id": "629",
"system": "neurological",
"primary_complaint": "cough",
"confirmed_symptoms": [
{
"value": "cough",
"status": "Yes",
"when": "last week"
},
{
"value": "diarrhea",
"status": "Yes"
},
{
"value": "headache",
"status": "Yes",
... | You are an advanced Clinical Natural Language Processing (NLP) system. Your task is to analyze the provided doctor-patient conversation transcript and extract all structured medical information into a comprehensive JSON object.
### Extraction Guidelines
Parse the dialogue to identify and categorize the following clini... | {
"Chief Complaint": "Cough and diarrhea",
"History of Present Illness": {
"Cough": {
"Onset": "5 days ago",
"Progression": "Worsened initially, currently improving",
"Characteristics": "Dry",
"Sputum": "None",
"Hemoptysis": false
},
"Diarrhea": {
"Onset": "2 days ago... |
N/A | {
"dialogue_id": "1004",
"system": "respiratory",
"primary_complaint": "swelling",
"confirmed_symptoms": [
{
"value": "swelling",
"status": "Yes",
"location": "calf(R)"
},
{
"value": "tenderness",
"status": "Yes",
"location": "calf(R)"
},
{
"value": ... | You are an advanced Clinical Natural Language Processing (NLP) system. Your task is to analyze the provided doctor-patient conversation transcript and extract all relevant structured medical information.
### Instructions
1. **Analyze the Transcript**: Read the conversation carefully to identify symptoms, history, medi... | {
"chief_complaint": "Hemoptysis and right calf swelling with tenderness",
"history_of_present_illness": {
"hemoptysis": {
"onset": "This morning (approximately 2 hours prior to visit)",
"frequency": "3 episodes",
"characteristics": "Spots of blood on tissue upon coughing",
"prior_cough"... |
N/A | {
"dialogue_id": "334",
"system": "respiratory",
"primary_complaint": "dyspnea",
"confirmed_symptoms": [
{
"value": "dyspnea",
"status": "Yes",
"when": "couple of days ago",
"other": "couple of days ago patient just started to not feel so well"
},
{
"value": "cough",
... | You are an advanced Clinical NLP System. Your task is to extract all structured medical information from the provided doctor-patient conversation transcript and return a single, comprehensive JSON object.
### Instructions
1. **Analyze the Transcript**: Read the conversation below carefully to identify symptoms, histor... | {
"chief_complaint": "Difficulty breathing and cough",
"history_of_present_illness": {
"onset": "Approximately 3 days ago (patient estimates Monday)",
"duration": "3 days",
"severity_and_progression": "Progressive worsening; patient reports today is the worst day",
"associated_symptoms": {
"co... |
N/A | {
"dialogue_id": "602",
"system": "respiratory",
"primary_complaint": "pressure",
"confirmed_symptoms": [
{
"value": "pressure",
"status": "Yes",
"when": "two week ago",
"location": "head, jaw, face"
},
{
"value": "nasal congestion",
"status": "Yes",
"progre... | You are an advanced Clinical Natural Language Processing (NLP) system. Your task is to analyze the provided doctor-patient conversation transcript and extract all structured medical information into a single, comprehensive JSON object.
### Extraction Guidelines
Parse the dialogue to identify and categorize the followi... | {
"chief_complaint": {
"primary_reason": "Nasal congestion and facial pressure",
"patient_description": "A lot of congestion in my nose and then I've also been feeling a lot of pressure in my head as well as just, around my jaws and my face overall"
},
"history_of_present_illness": {
"onset": "2 weeks... |
End of preview. Expand in Data Studio
The heliosbrahma/mental_health_chatbot_dataset is a specialized resource designed to support the development of conversational AI agents capable of providing mental health information and support. The dataset consists of English-language dialogue pairs that simulate interactions between individuals seeking help and healthcare providers.
Content within the dataset was derived from established medical resources, including WebMD, Mayo Clinic, and Healthline, alongside frequently asked questions from reputable online health platforms. To ensure suitability for public use and model training, all entries have undergone rigorous preprocessing. This includes the removal of personally identifiable information (PII) and the elimination of extraneous characters, ensuring that the data remains anonymized while retaining clinical relevance.
Structurally, each data instance contains a single field labeled text, which houses the complete exchange: a user inquiry followed by a corresponding professional response. The primary objective behind curating this collection is to facilitate the fine-tuning of artificial intelligence models. These models aim to deliver accessible, immediate, and non-judgmental support, serving as a supplementary tool to human care, particularly during times of distress. By leveraging verified medical content, the dataset supports the creation of chatbots that can offer reliable information while maintaining a calm and empathetic tone.
- Downloads last month
- 36