Audit to Forget: A Unified Method to Revoke Patients’ Private Data in Intelligent Healthcare

Event Start
Event End
Building 2, Level 5, Room 5209


Revoking personal private data is one of the basic human rights, which has already been sheltered by several privacy-preserving laws in many countries. As data science, machine learning, and deep learning techniques continue to advance, such right is often overlooked or infringed upon due to the increasing collection and use of patient data for model training. The prevalence of such violations in intelligent healthcare highlights the need for technology to comply with laws, regulations, and privacy principles to ensure that innovation serves the common good. In order to secure patients’ right to be forgotten, we proposed a novel solution by using auditing to guide the forgetting process, where auditing means determining whether a dataset has been used to train the model and forgetting requires the information of a query dataset to be forgotten from the target model. We unified these two tasks by introducing a new approach called knowledge purification. To implement our solution, we developed an audit to forget software (AFS), which is able to evaluate and revoke patients’ private data from pre-trained deep learning models. To demonstrate the generality of AFS, we applied it to four tasks based on four datasets, including the MNIST dataset, the PathMNIST dataset, the COVIDx dataset, and the ASD dataset, with different data sizes and various architectures of deep learning networks. Our results demonstrate the usability of AFS and its application potential in real-world intelligent healthcare to enhance privacy protection and data revocation rights.

Brief Biography

Juexiao Zhou is an M.S./Ph.D. candidate in the KAUST Structural and Functional Bioinformatics research group under the supervision of Professor Xin Gao. Before joining KAUST, Juexiao majored in Bioinformatics and graduated from Southern University of Science and Technology (SUSTech), located in Shenzhen, Guangdong, China.

Juexiao focuses on the research of the interdisciplinary subjects, including Biology and Computer Science. In SUSTech, He worked on RNA regulation, protein functions, biochemical pathways, and tumor reprogramming. After joining KAUST, he turned to the application of deep learning in Biology and medical imaging. Visit his website for more information.

Contact Person