CLASSify- A Web-based Tool for Machine Learning

Clinicians often produce large amounts of data from various sources, such as patient metrics, drug components, and treatment outcomes. Classifying this data is key to gaining insights or identifying trends that could lead to improvements in patient care. However, this is often a time-consuming process. While clinicians are interested in using AI to address this challenge, they may lack the technical skillsets needed to leverage the technology. CAAI developed CLASSify to make machine learning (ML) easier and more accessible for clinicians and other users who may not have technical or programming backgrounds. CLASSify provides an easy-to-use, web-based platform to train and evaluate ML classification models on any tabular data. Users can simply upload their dataset to the site using the default settings, or selecting their specific training parameters. The job will be sent off to train all chosen models and provide results in the form of tables and visualizations. CLASSify also provides options for synthetic data generation to bolster imbalanced class labels or create entirely new datasets. Additionally, CLASSify determines explainability scores that provide insight into which features of the data are most important to the models’ predictions. As a self-service, all-in-one ML tool, CLASSify allows clinicians to effortlessly compare models, gather results, and download any generated artifacts for later use.

Why Use CLASSify?

CLASSify provides a variety of machine learning models for tabular classification problems. This means that the data used should take the form of a table, with rows representing individual observations and columns representing features, and there is some classification you want to train a model to predict for those types of observations. This classification can be binary (only two classes) or multiclass (more than two classes). CLASSify is flexible to a variety of data types, missing data, and dataset sizes. CLASSify allows you to easily train and evaluate models without any coding expertise. The training jobs you can run are secure and highly customizable, and any artifacts or models created can be downloaded for further use. You can also use the interface to generate synthetic data and produce explainability charts to learn more about your data. You can easily re-run jobs with modified parameters to compare results and upload new data for further model evaluation and prediction.

More Information

Additional information, including a technical overview and user guide documentation can be found here: https://hub.ai.uky.edu/classify/

Accessing CLASSify

CLASSify is available for free on an individual basis on CAAI’s self-service tool website. Before you can get started, you must be granted the necessary permissions from a CAAI Administrator. Please contact us for access or submit our collaboration intake form here.

Instructional Video

A video explaining CLASSify and how it is used can be found here.

Citation

A paper detailing the development and usage of this tool was submitted and accepted to the American Medical Informatics Association (AMIA) in 2023. This paper can be found here: https://arxiv.org/abs/2310.03618