Authors: Yujiang Wang, Anshul Thakur, Mingzhi Dong, Pingchuan Ma, Stavros Petridis, Li Shang, Tingting Zhu, David A. Clifton

Published on: May 05, 2023

Impact Score: 8.45

Arxiv code: Arxiv:2305.03711

Summary

• What is new: This research introduces the use of dataset condensation (DC) to anonymise and share medical data efficiently for AI research.
• Why this is important: The need for anonymisation of sensitive clinical data and free knowledge flow for AI studies in healthcare.
• What the research proposes: Implementing dataset condensation to create condensed data that maintains the utility for AI models while preserving privacy.
• Results: Experimental results show that the DC approach effectively abstracts clinical records for AI research across three different healthcare datasets.

Technical Details

Technological frameworks used: Dataset condensation methods for deep learning

Models used: Deep neural networks (DNNs)

Data used: Electrical healthcare records (EHRs)

Potential Impact

Healthcare providers, AI healthcare startups, medical data analytics companies, and patient privacy-focused organizations.

Want to implement this idea in a business?

We have generated a startup concept here: HealthSynth.