A deep learning method to detect personal and family history of suicidal thoughts and behaviors.
This repository maintains the code for detection of personal and family history of suicidal thoughts and behaviours in clinical notes. Pre-trained Bio_Clinical_BERT and GatorTron BERT models are implemented for fine-tunning (using annotated corpus) and classification.
cd (change directory) to location of cloned repository
Run the command python3 -m venv stb_env
Run the command source stb_env/bin/activate in terminal
Run the command deactivate in terminal.
If using database for data input/output, provide connection details in mysql_connection.py or sql_connection.py files
Run various python scripts:
Example Command: python fsh_BioClinicalBERT_classifier.py
Example Command: python fsh_GatorTron_classifier.py
Example Command: python sh_BioClinicalBERT_classifier.py
Example Command: python sh_GatorTron_classifier.py