Download This Paper
Preprints with The Lancet is a collaboration between The Lancet Group of journals and SSRN to facilitate the open sharing of preprints for early engagement, community comment, and collaboration. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early-stage research papers that have not been peer-reviewed. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. The findings should not be used for clinical or public health decision-making or presented without highlighting these facts. For more information, please see the FAQs.
Digital Fd: Enhancing Privacy Protection and Facilitating Auxiliary Diagnosis of Ocular Diseases
21 Pages Posted: 17 May 2024
More...Abstract
Biometric data extracted from facial images holds significant commercial value, yet raises substantial security concerns. Inspired by the discussions of Yang et al. and Meeus et al. in their respective publications in Nature Medicine (2023), which discuss the effectiveness of digital masks (DM) in preserving patients' privacy, we introduce Digital FaceDefender (Digital FD) as an innovative strategy aimed not only at safeguarding patient privacy but also at facilitating auxiliary diagnosis. Following the generation of Asian-face virtual avatars, 52 landmarks from bothvirtual avatars and patients were captured for subsequent image fusion using Google's MediaPipe library. Affine Transformation of the Eye Region between patients and avatars was performed to create preliminary fusion images, followed by color correction and Gaussian blur to enhance final fusion effects. Verification experiments employing ArcFace and DECA methodologies were conducted on a sample of 50 individuals, each represented by five images sourced from the CASIA-FaceV5 dataset. Furthermore, two auxiliary diagnostics models were constructed using a dataset comprising 1163 healthy individuals. Utilizing the Digital FD framework, the Digital FD platform was established and made accessible. In contrast to conventional digital masks characterized by extensive coverage, Digital FD offers a natural aesthetic that not only safeguards patients' privacy but also fosters empathy between healthcare providers and patients, concurrently facilitating auxiliary diagnoses. Additionally, the results of the similarity score and Rank-1 accuracy analysis indicated a reduced likelihood of reidentification when employing Digital FD. In summary, our findings exhibit the efficacy of Digital FD in preserving patients' privacy and facilitating auxiliary diagnoses.
Funding: Our research was partially supported by National Natural Science Foundation of China grants No.12231017, 72171216, 71921001, and 71991474, and Innovative development funds of Anhui Province Federation of Social Sciences （No.2022CX081）.
Declaration of Interest: The authors declare no competing interests.
Ethical Approval: The research protocol and ethical review process for this study received approval from the Institutional Review Board/Ethics Committee of the Fourth Affiliated Hospital of Harbin Medical University (2023-Ethics Review-54). Informed consent was obtained from all individuals participating in the study.
Keywords: Digital FD, privacy protection, auxiliary diagnosis, Artificial Intelligence, ocular diseases
Suggested Citation: Suggested Citation