Skip to content

Update overview.md #3

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
alexkarargyris wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Update overview.md #3

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
30 changes: 21 additions & 9 deletions docs/overview.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,17 +1,29 @@
# Tutorial: How to run an AI study in a clinical setting using GaNDLF and OpenFL on MedPerf

This section presents the general audience of the entire tutorial, the goals for each audience, and also how they will be achieved.
The aim of this tutorial is to provide a detailed description on running a clinical AI study across multiple sites using GaNDLF and OpenFL on MedPerf. At the end of the tutorial, depending on your organization’s role, you should be comfortable with understanding what a clinical AI study is, how to curate medical data, how to prepare data for ingestion by the machine learning (ML) experiments, how to train ML models on multiple distributed sites using Federated Learning, and finally how to perform evaluation on these sites using Federated Evaluation.

## General audience
*Important: At its current stage this tutorial does not take into consideration the health IT system complexities (e.g. databases, storage, networking) that affect mainly data extraction. It tries to abstact it from these complexities and focus on the machine learning aspects of the clincal AI study.*

- Hospital IT Experts (i.e. setup MedPerf)
- Clinical Experts (i.e. data curation, data preparation: specifically annotation)
- ML Engineeris (i.e. data preparation, model dev, training, evaluation)
## Roles

### Healthcare IT expert
In the context of this tutorial a healthcare IT expert has the following responsibilities:
* Understands healthcare organization network specifications and requirements,
* Installs and sets up containerized software on a local server,
* Understands and provides secure outgoing network communication with an external orchestrator server.

## Goals for each audience
- **Hospital IT Experts**: Understand how to set up MedPerf (i.e. installation, testing, creating accounts on MedPerf, identify technical issues, reporting issues)
- **Clinical Experts**: Define the clinical problem, perform the annotations, and make analysis
- **ML Engineers**: Prepare the ML pipeline (i.e. design data preparation MLCube, FL training MLCube, FL validation MLCube, implement evaluation metrics MLCube)
### Clinical expert
In the context of this tutorial, a clinical expert has the following responsibilities:
* Defines the clinical study: a clinical problem, expected input and output of the AI model, data acquisition (inclusion & exclusion criteria), annotation protocol,
* Provides clinical expertise to the other stakeholders (i.e. healthcare IT expert, machine learning engineer),
* Provides data annotations according to the annotation protocol,
* Provides feedback and issues to other stakeholders

### Machine learning engineer
In the context of this tutorial, a machine learning engineer has the following responsibilities:
* Communicates with other stakeholders
* Prepares ML pipeline and experiments
* Helps setup necessary tools related to this tutorial

## Roadmap to achieving the goals
- *Provide source code examples through a use case: BrATS*
Expand Down