Trial Checker

Introduction

TrialChecker is a model designed to assess the "reasonable consideration" of cancer clinical trials based on patient summaries. It utilizes the RoBERTa architecture for text classification tasks in the medical domain, specifically focusing on clinical trials and cancer-related data.

Architecture

The model is built on the FacebookAI/roberta-large base model, leveraging the capabilities of the Transformers library. It is tailored for text classification tasks, particularly in the medical and clinical trial contexts. The model employs embedding techniques to understand and process English text data related to clinical trials and cancer.

Training

TrialChecker was trained using data relevant to medical and clinical trials, focusing on embedding techniques to enhance the model's understanding of complex medical texts. The training process involves fine-tuning the RoBERTa architecture to improve its performance in assessing clinical trial suitability based on patient summaries.

Guide: Running Locally

To run the TrialChecker model locally, follow these steps:

1. Install the Transformers Library:
   Make sure you have the Transformers library installed. You can do this via pip:

   pip install transformers
   

2. Download the Model:
   Clone or download the model files from the Hugging Face repository:

   git clone https://huggingface.co/ksg-dfci/TrialChecker
   

3. Load the Model:
   Use the Transformers library to load the model:

   from transformers import AutoModelForSequenceClassification, AutoTokenizer
   
   tokenizer = AutoTokenizer.from_pretrained("ksg-dfci/TrialChecker")
   model = AutoModelForSequenceClassification.from_pretrained("ksg-dfci/TrialChecker")
   

4. Inference:
   Prepare your input data and run inference using the model:

   inputs = tokenizer("Your patient summary here", return_tensors="pt")
   outputs = model(**inputs)
   

Consider using cloud GPUs such as AWS EC2, Google Cloud, or Azure to speed up the process if running on large datasets or requiring faster inference.

License

TrialChecker is licensed under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0). This license allows others to use, share, and adapt the model for non-commercial purposes, provided proper attribution is given.