CoolFace
Datasetpublic

monfortbrian/lung_cancer_5K.jsonl

Lung Cancer Dataset 🫁 A curated dataset of prompt–completion pairs designed for fine-tuning Large Language Models (LLMs) on lung cancer diagnostics.The dataset contains 5,000 rows of text pairs prepared for medical AI research, clinical assistants, and healthcare copilots. 📊 Dataset Overview Size: 5,000 prompt–completion pairs Format: JSONL, CSV Domain: Lung Cancer (diagnosis, symptoms, treatment, follow-up) Use Case: Training LLMs for Doctor Copilot and… See the full description on the dataset page: https://huggingface.co/datasets/monfortbrian/lung_cancer_5K.jsonl.

sourceHugging Facemitupdated 1y agoView on Hugging Face
1likes24downloads
Dataset Card

Lung Cancer Dataset 🫁

A curated dataset of prompt–completion pairs designed for fine-tuning Large Language Models (LLMs) on lung cancer diagnostics. The dataset contains 5,000 rows of text pairs prepared for medical AI research, clinical assistants, and healthcare copilots.

📊 Dataset Overview

  • —Size: 5,000 prompt–completion pairs
  • —Format: JSONL, CSV
  • —Domain: Lung Cancer (diagnosis, symptoms, treatment, follow-up)
  • —Use Case: Training LLMs for Doctor Copilot and clinical diagnostic support

🛠️ Data Source & Preparation

  • —Collection: Derived from internet-sourced content, carefully paraphrased and curated.
  • —Processing: Cleaned, normalized, and anonymized to remove sensitive information.
  • —Structure: Optimized for PEFT / LoRA fine-tuning, with balanced prompt–completion alignment.

🚀 Applications

This dataset is intended for:

  • —Fine-tuning LLMs for fast, domain-specific medical inference
  • —Building diagnostic copilots for doctors and healthcare professionals
  • —Research into AI-assisted cancer detection & treatment planning

📂 File Formats

  • —lung_cancer_dataset.jsonl → JSON Lines format (recommended for training)
  • —lung_cancer_dataset.csv → CSV format for exploration and preprocessing

⚠️ Disclaimer

This dataset is for research and educational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment.

💡 Inspiration

This dataset is part of the Doctor Copilot MVP, aiming to create safe, reliable, and efficient AI-assisted medical systems for early detection and better patient outcomes.