Multi-modal dataset creation for federated learning with DICOM-structured reports - Summary - MDSpire

Multi-modal dataset creation for federated learning with DICOM-structured reports

  • By

  • Malte Tölle

  • Lukas Burger

  • Halvar Kelm

  • Florian André

  • Peter Bannas

  • Gerhard Diller

  • Norbert Frey

  • Philipp Garthe

  • Stefan Groß

  • Anja Hennemuth

  • Lars Kaderali

  • Nina Krüger

  • Andreas Leha

  • Simon Martin

  • Alexander Meyer

  • Eike Nagel

  • Stefan Orwat

  • Clemens Scherer

  • Moritz Seiffert

  • Jan Moritz Seliger

  • Stefan Simm

  • Tim Friede

  • Tim Seidler

  • Sandy Engelhardt

  • February 3, 2025

  • 0 min

Share

Objective:

To develop a method for harmonizing multi-modal medical data using DICOM standards for federated learning, emphasizing its significance in enhancing data sharing and model training.

Key Findings:
  • DICOM standard ensures harmonized data representation across institutions, crucial for effective federated learning.
  • The developed dashboard allows for iterative filtering of multi-modal data, enhancing user experience.
  • Interoperability is maintained with the ability to query DICOM data using SQL, supporting diverse analytical needs.
Interpretation:

The study demonstrates that leveraging DICOM standards can facilitate the creation of large-scale, harmonized datasets necessary for effective federated learning in multi-modal medical applications, potentially transforming data sharing practices.

Limitations:
  • Dependence on adherence to DICOM templates across institutions, which may vary in compliance.
  • Privacy regulations may still limit data sharing capabilities, impacting the scalability of the approach.
Conclusion:

The proposed method enhances the feasibility of federated learning in multi-modal medical scenarios by ensuring data harmonization and providing user-friendly tools for cohort selection, which is vital for advancing collaborative research.

Original Source(s)

Multi-modal dataset creation for federated learning with DICOM-structured reports

  • by Malte Tölle, Lukas Burger, Halvar Kelm, Florian André, Peter Bannas, Gerhard Diller, Norbert Frey, Philipp Garthe, Stefan Groß, Anja Hennemuth, Lars Kaderali, Nina Krüger, Andreas Leha, Simon Martin, Alexander Meyer, Eike Nagel, Stefan Orwat, Clemens Scherer, Moritz Seiffert, Jan Moritz Seliger, Stefan Simm, Tim Friede, Tim Seidler, Sandy Engelhardt

  • February 3, 2025

Related Content

AI Supports Glaucoma Surgical Planning

Predictive models identify high-risk patients and support earlier treatment decisions.

Multimodal human–computer interaction in interventional radiology and surgery: a systematic literature review

Cataract Surgeons Gain a New Lens Tool

An extended depth of focus intraocular lens approved by the FDA is designed to improve visual range while maintaining contrast sensitivity in cataract surgery.