Skip to content

michael-s-yao/PennMedBiobankAIDeployment

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Deployment Strategies for Machine Learning Algorithms in Patient Healthcare Domains: An Imaging Perspective

LICENSE CONTACT CONTACT

Despite recent advancements in machine learning and artificial intelligence for applications to patient healthcare, real hospital settings have experienced few of the potential benefits and improvements to clinical medicine. To facilitate clinical adaptation of methods in machine learning, we propose DPSP: a standardized framework for step-by-step deployment that focuses on four key components: Data acquisition; Problem identification; Stakeholder alignment; and Pipeline integration. We leverage recent literature and empirical evidence in radiologic imaging applications to justify our approach, and offer discussion to help other research groups and hospital practices leverage machine learning to improve patient care.

Installation

To install and run our code, first clone the PennMedBiobankAIDeployment repository.

git clone https://github.com/michael-s-yao/PennMedBiobankAIDeployment
cd PennMedBiobankAIDeployment

Next, create a virtual environment and install the relevant dependencies.

python -m venv env
source env/bin/activate
pip install -r requirements.txt

Quantitative figures presented in our work can be easily reproduced by running the following commands:

python PMBB.py --use_sans_serif
python distribution_shift.py

Contact

Questions and comments are welcome. Suggests can be submitted through Github issues. Contact information is linked below.

Allison Chae

Michael Yao

Citation

When available, relevant citation information will be added in a future commit.

License

This repository is MIT licensed (see LICENSE).

About

Codebase for Deployment Strategies for Machine Learning Algorithms in Patient Healthcare Domains: An Imaging Perspective

Topics

Resources

License

Stars

Watchers

Forks

Languages