Kevin Thandiackal
Papers
1
Total Citations
4
H-Index
1
About
Kevin Thandiackal is a researcher at the forefront of privacy-preserving machine learning and audio signal processing, with a particular focus on respiratory sound analysis. His work addresses the critical challenge of developing robust AI systems that can learn incrementally from sensitive medical data without compromising patient confidentiality. In his most-cited paper, "Privacy preserving synthetic respiratory sounds for class incremental learning" (2021, 4 citations), Thandiackal pioneered methods to generate synthetic respiratory sounds that retain diagnostic utility while protecting individual privacy. This contribution is vital for enabling continuous learning in healthcare AI, where models must adapt to new disease classes over time without accessing original recordings. His research bridges the gap between data privacy regulations and the need for scalable, adaptive machine learning in clinical settings. By tackling the intersection of synthetic data generation, class incremental learning, and audio-based diagnostics, Thandiackal is helping to build a future where AI can learn from sensitive health data ethically and effectively—a crucial step toward deploying trustworthy diagnostic tools in real-world medical environments.
Research Focus
Key Achievements
Top Papers
- 1