MEDomicsLab-docs
V1
V1
  • πŸ‘‹Welcome!
  • πŸ‘ŠQuick start
  • πŸ‘€Overview
  • 🏫Tutorials
    • πŸ”΅Design
      • Extraction Module
        • Image Extraction Page
        • Text Extraction Page
        • Time Series Extraction Page
        • MEDimage
      • Input Module
        • Feature Reduction Tool
        • MEDprofiles
          • MEDprofiles Viewer
      • Exploratory Module
    • 🟠Development
      • Learning Module
      • Evaluation Module
    • 🟒Deployment
      • Application Module
  • πŸ†•New features
  • Upcoming features
  • πŸ’»Contributing
    • Our coding standards
    • How to push my modification ?
  • πŸ€•Troubleshooting
  • ❓FAQ
  • πŸ€“About us
  • Important Links
    • Official Website
    • πŸ“”Release Notes
    • πŸ₯²Known Issues
    • 😎Project Board
    • 🧬Physionet
  • MEDIA
    • βš›οΈMEDomics
    • πŸ‘ΎDiscord
    • 😺Github
    • πŸ“ΊYouTube
  • Forms
    • πŸ—£οΈContact us
    • πŸ“Report an issue
Powered by GitBook
On this page
  • Superset
  • MEDfl
  • MED3pa
Edit on GitHub

Upcoming features

Explore the upcoming implementations in the next releases.

PreviousNew featuresNextContributing

Last updated 6 months ago

Superset

is an open-source modern data exploration and visualization platform, loaded with options that make it easy for users of all skill sets to explore and visualize their data, from simple line charts to highly detailed ones. MEDomicsLab will integrate direct access to Superset, eliminating the need for multiple installations and making MEDomicsLab a comprehensive, all-in-one solution for modeling and visualizing heterogeneous data in medicine.

MEDfl

Key Aspects of the Federated Learning Module:

  • Decentralized Training: Models are trained across multiple simulated nodes without transferring raw data.

  • Privacy Preservation: Utilizing techniques like differential privacy to ensure data confidentiality.

  • Hyperparameter Optimization: Tools to automatically tune and optimize model hyperparameters for improved performance.

  • Transfer Learning: Allows the user to use pre-trained models to initialize the central server, enhancing model performance.

MED3pa

Key Functionalities

  • Covariate Shift Detection: Utilizing the Detectron sub-package, MED3pa can identify significant shifts in data distributions that might affect the model’s predictions. This feature is crucial for applications such as healthcare, where early detection of shifts can prevent erroneous decisions.

  • Uncertainty and Confidence Estimation: Through the med3pa subpackage, the package measures the uncertainty and predictive confidence at both individual and group levels. This helps in understanding the reliability of model predictions and in making informed decisions based on model outputs.

  • Identification of Problematic Profiles: MED3pa analyzes data profiles that consistently lead to poor model performance. This capability allows developers to refine training datasets or retrain models to handle these edge cases effectively.

In the upcoming releases, MEDomicsLab will offer a graphical implementation of the package . This module, part of the development layer, represents the third major component in MEDomicsLab’s toolset, enabling simulated federated learning to support collaborative model training across multiple sites without sharing raw data. By facilitating decentralized training, this module enhances data privacy and security, empowering researchers and developers to build, refine, and deploy federated learning models directly within the platform.

MEDomicsLab will also offer a graphical implementation of the package. It is designed to address critical challenges in deploying machine learning models, particularly focusing on the robustness and reliability of models under real-world conditions. It provides comprehensive tools for evaluating model stability and performance in the face of covariate shifts, uncertainty, and problematic data profiles.

Federated Learning
MEDfl
MED3pa
Apache Superset