Privacy Preservation Technique in Cardiovascular Disease Prediction Using Federated Learning Model

#collaboration #machine-learning #LTU #privacy #learning #organizations #confidentiality
Share

Abstract: Cardiovascular disease (CVD) is one of the leading causes of death worldwide, making early and accurate prediction essential for improving patient outcomes. Traditional machine learning approaches require centralized medical data, which raises concerns about patient privacy and security. This project proposes a privacy-preserving cardiovascular disease prediction system using a Federated Learning (FL) model, where multiple healthcare institutions collaboratively train a shared model without exchanging sensitive patient data. Each institution trains the model locally, and only encrypted model updates are shared with a central server for aggregation. This approach protects patient confidentiality while achieving prediction performance comparable to centralized models. The proposed framework enhances privacy, security, scalability, and collaboration among healthcare organizations, making it a reliable solution for intelligent and secure cardiovascular disease prediction.


  Date and Time

  Location

  Hosts

  Registration


  • Add_To_Calendar_icon Add Event to Calendar

Loading virtual attendance info...

  • Contact Event Hosts
  • Starts 30 July 2026 04:00 AM UTC
  • Ends 07 August 2026 04:00 AM UTC
  • No Admission Charge

  Speakers

Shirisha of Lawrence Technological University

Topic:

Privacy Preservation Technique in Cardiovascular Disease Prediction Using Federated Learning Model

Cardiovascular disease (CVD) is one of the leading causes of death worldwide, making early and accurate prediction essential for improving patient outcomes. Traditional machine learning approaches require centralized medical data, which raises concerns about patient privacy and security. This project proposes a privacy-preserving cardiovascular disease prediction system using a Federated Learning (FL) model, where multiple healthcare institutions collaboratively train a shared model without exchanging sensitive patient data. Each institution trains the model locally, and only encrypted model updates are shared with a central server for aggregation. This approach protects patient confidentiality while achieving prediction performance comparable to centralized models. The proposed framework enhances privacy, security, scalability, and collaboration among healthcare organizations, making it a reliable solution for intelligent and secure cardiovascular disease prediction.

Biography:

Shirisha completed her Masters from Lawrence Technological University in May 2026. 

Email:

Address:21000 West Ten Mile Road, Southfield, MI, 48075-1058, Southfield, Michigan, United States, 48075




Agenda

10.00 am - Introduction about the speaker

10.05 am - Talk 

10.45 am - Q&A

11.00 am - Conclusion