Ay, Seha
Cardei, Michael
Meyer, Anne-Marie
Zhang, Wei
Topaloglu, Umit
Funding for this research was provided by:
Bioinformatics Shared Resource of the Wake Forest Baptist Comprehensive Cancer Center’s NCI Cancer Center Support Grant (P30CA012197, P30CA012197)
National Center for Advancing Translational Sciences, National Institutes of Health (UL1TR001420, UL1TR001420)
National Cancer Institute (3R01CA272627-01A1S1)
Wake Forest University
Article History
Received: 14 June 2023
Revised: 2 February 2024
Accepted: 19 February 2024
First Online: 28 February 2024
Declarations
:
: The research presented in this manuscript titled “Improving Equity in Deep Learning Medical Applications with the Gerchberg-Saxton Algorithm” utilizes the MIMIC-III dataset, which is created by the Massachusetts Institute of Technology (MIT) and publicly available at physionet.org. The MIMIC-III dataset consists of de-identified human vital signs data collected from patients admitted to the Beth Israel Deaconess Medical Center.This research study does not involve the subjecting of animals or humans to any interventions or procedures. Instead, it relies solely on the analysis of the pre-existing MIMIC-III dataset for scientific investigation. The dataset has been made publicly available by MIT for research purposes, and all identifying information has been anonymized to ensure privacy and confidentiality.As the MIMIC-III dataset is openly accessible and contains de-identified information, the research team acknowledges and adheres to the following ethical considerations:1. Confidentiality: The research team will strictly adhere to data protection and confidentiality protocols. The dataset utilized in this study has been previously de-identified to safeguard the privacy of the individuals whose data is included.2. Informed Consent: Since no new data will be collected from individuals for this study, obtaining informed consent is not applicable. The MIMIC-III dataset has been released with the understanding that it will be used for scientific research purposes while maintaining confidentiality.3. Beneficence and Non-Maleficence: The study will be conducted with the principles of beneficence and non-maleficence in mind. All analyses will be carried out responsibly, aiming to generate valuable scientific insights while minimizing any potential harm or negative consequences.4. Compliance with Regulations: This research complies with all relevant laws, regulations, and guidelines pertaining to the use of publicly available datasets. The study adheres to the policies and terms of use set forth by the custodians of the MIMIC-III dataset.5. Acknowledgement: The custodians of the MIMIC-III dataset, including the Massachusetts Institute of Technology, will be duly acknowledged for their efforts in data collection and sharing. Proper citation and attribution will be provided in accordance with the dataset’s documentation and usage guidelines.By utilizing the MIMIC-III dataset, this research contributes to advancing knowledge in the field of bias mitigation in AI-powered healthcare solutions without subjecting any animals or humans to additional procedures or interventions. The findings presented in this manuscript are solely derived from the analysis of the publicly available dataset, adhering to ethical principles and regulations.
: The authors declare no competing interests.