| Abstract |
The gradual development of smart city, healthcare, and health data collection has paved the path for federated learning (FL)-based heart disease detection. In 2021, cardiovascular diseases claimed about twenty million deaths and accounted for more than a third of all the mortality in the world. For most preventable heart diseases, a proactive approach to detection and management bears high importance. FL allows the collaborative enhancement of an intelligent system by integrating locally extracted knowledge. This provides an opportunity for advancements in heart disease detection systems without the requirement of centralized patient data. After evaluating 17 studies that were published between the years 2017 and 2023, this survey discusses the state-of-the-art practices utilized under FL for heart disease detection. Novel techniques used by the researchers to create FL models for detecting heart diseases are also presented. The study highlights the potential gains of using FL including the increased accuracy and the generalizability of developed models. Ethical concerns such as issues in privacy, data governance, and heterogeneity have also been included. Overall, this study is highly informative on the advances of FL-based heart disease detection and the problems governing the healthcare domain in smart city applications. © 2025 selection and editorial matter, Diptendu Sinha Roy, Mir Wajahat Hussain, K. Hemant Kumar Reddy, Deepak Gupta; individual chapters, the contributors. |
