MDSD4Health.com is the website which houses the Machine Learning Model & Dataset Disclosure (MDSD) for Healthcare and Public Health curriculum. The curriculum covers topics in machine learning, bias and fairness in automated systems, and disclosure methods and mediums.

MDSD4Health was developed by Eden Shaveet as a capstone deliverable toward the completion of her Master of Science degree in health informatics and analytics (HIA) at Tufts University School of Medicine in 2022. This work was precepted by Vivian Neilley, MS-HIA, Product Manager at Google Cloud, Healthcare & Life Sciences.

See a more exhaustive list of the people who this work could not exist without here!

Great question! Disclosure practices within the realm of machine learning (ML) refer to deliberate communication of an ML model's origins, including (but not limited to) its known development limitations, performance metrics, intended uses, and training and testing dataset composition. The development of robust disclosure methods and mediums have been proposed as a central strategy in the pursuit of achieving algorithmic transparency in healthcare and public health contexts.

For more on this, see Module 2: The Role of Disclosure.

Thank you for helping us make this curriculum better! You can report a problem or suggest a revision to the MDSD4Health curriculum here.

All original and curated MDSD4Health content is freely available on the MDSD4Health site as well as via any original publisher's communication channels. We encourage you to review our Learning Guide for a high-level summary of each content module, including key concepts and associated time completion estimates, to determine what content integration is most feasible and applicable to your own course offerings.

If you decide to integrate curated materials, please always cite the original content creators (e.g. "Crash Course Computer Science" or "Google News Initiative," etc.). If you decide to integrate our interpretations of these resources, original MDSD4Health videos, or our Colab exercises, please also cite us: Shaveet, E. (2022). Model and Dataset Disclosures for Healthcare & Public Health. MDSD4Health. https://www.mdsd4health.com

We welcome and encourage involvement and contributions from anyone interested in communicating the importance of MDSDs via free-access education. Here are a few ways to get involved:

• Suggest changes or additions to a module's content via the Feedback Form.

• Review and revise our coding exercises on GitHub.

• Use our Learning Guide to integrate all (or some) of the content into your own coursework.

  • • If you decide to integrate MDSD4Health content into a course, please consider sharing your materials (slides, syllabus, etc.) with us via email to share as examples for others!