Dr. Didem Stark

Didem Stark
Didem Stark is interested in understanding bias and fairness in machine learning models used for clinical neuroimaging data. Using various fairness metrics, her research investigates hidden biases that may disadvantage minority groups. More importantly, she focuses on how debiasing methods perform across diverse subpopulations. As AI becomes increasingly integrated into healthcare decision-making, her research is ultimately aiming to ensure that these powerful tools benefit all members of society equally. With plans to apply for the Walter Benjamin funding programme, she seeks to contribute to ethical AI in medicine, particularly in clinical neuroscience.
In medical AI applications, fairness is fundamental to ensuring equitable benefits for everyone in society.
01
Rane, R.P., Kim, J., Umesha, A., Stark, D., Schulz, MA., Ritter, K.

DeepRepViz: Identifying Potential Confounders in Deep Learning Model Predictions

Oct 03, 2024 | MICCAI 2024
02
Klingenberg, M., Stark, D., Eitel, F., Ritter, K., & Alzheimer’s Disease Neuroimaging Initiative

MRI Image Registration Considerably Improves CNN-Based Disease Classification

Sep 21, 2021 | Machine Learning in Clinical Neuroimaging 2021
03
Stark, D., & Ritter, K.

AIM and Gender Aspects

Jan 01, 2020 | Artificial Intelligence in Medicine