Manara - Qatar Research Repository
Browse

Federated Learning Using Imbalanced Medical Image Dataset

Download (876.39 kB)
thesis
submitted on 2024-10-30, 10:12 and posted on 2024-10-30, 10:12 authored by Khalid Mahmoud Mohammad Dolaat
Federated learning (FL) is a deep-learning framework developed for cases where data privacy protection is essential, due to security and privacy guidelines. Collaboration between independent institutions is critical to achieving better deep-learning model accuracy due to a single institution's shortage of information acquisition. In this study, we focus on medical image analysis of brain tumor dataset, using a federated learning framework. One of the drawbacks that come with FL is the global model accuracy impact as a result of the imbalanced distribution of dataset's samples and classes called non-independent and identically distributed (non-IID). FL global model accuracy degradation can be optimized by various approaches, either by manipulating the dataset using augmentation techniques and data sharing or optimizing the global model aggregation algorithm. Our method uses two augmentation techniques: Generative adversarial network (GAN) and Synthetic Minority Over-sampling Technique (SMOTE) to balance non-IID with the typical FedAvg aggregation algorithm to optimize global model accuracy. Global model accuracy using SMOTE and GAN outperform IID and non-IID with a few numbers of global model iterations using local imbalance or globally imbalanced datasets.

History

Language

  • English

Publication Year

  • 2022

License statement

© The author. The author has granted HBKU and Qatar Foundation a non-exclusive, worldwide, perpetual, irrevocable, royalty-free license to reproduce, display and distribute the manuscript in whole or in part in any form to be posted in digital or print format and made available to the public at no charge. Unless otherwise specified in the copyright statement or the metadata, all rights are reserved by the copyright holder. For permission to reuse content, please contact the author.

Institution affiliated with

  • Hamad Bin Khalifa University
  • College of Science and Engineering - HBKU

Degree Date

  • 2022

Degree Type

  • Master's

Advisors

M. Erbad Aiman Mohmood

Committee Members

Marco Agus ; Tanvir Alam ; Tareq Al-Ansari

Department/Program

College of Science & Engineering

Usage metrics

    College of Science and Engineering - HBKU

    Categories

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC