ArticlePDF Available


The decisions made by Artificial Intelligence (AI) systems are critical due to their growing usage in sensitive areas such as recruitment, criminal justice, and healthcare. It is dramatically significant to detect and measure AI bias to mitigate the effects of bias by making these systems more transparent, explainable, and auditable. In this study, we focus on gender bias to investigate the effect of gender imbalance in medical imaging dataset when applying AI models to detect Covid-19. We perform an analysis to measure gender bias in the diagnosis of medical imaging data using deep learning-based methods. We primarily examine the distribution of samples based on metadata and target labels. In the training phase, we conduct experiments to reveal that gender imbalance produces a biased model. For this purpose, we train a model using both a fully gender-balanced and immensely imbalanced dataset unique to a specific gender. To show that the inferences are generalizable, we apply several deep learning-based solutions including pre-trained models. We compare the performance of different models for exploring gender bias. We observe a significant difference in classification performances between trained models using the imbalanced dataset and balanced dataset in terms of gender. We confirm a similar tendency when using different deep learning methods. Consequently, our experimental results show that gender-imbalance in medical imaging data produces biased decisions in Covid-19 detection. In this study, we explore a gender bias in the deep learning aided Covid-19 diagnosis of the gender-unbalanced medical image data.
... Artificial Intelligence (AI) based solutions, in particular, deep learning and machine learning are increasingly used in several domains in healthcare such as identifying disease (Jaiswal et al., 2019), detecting bias in methods used to diagnose diseases (Dervisoglu et al., 2021), detecting of COVID19 from symptomatic information (Najar, 2021), etc. Also, there are many opportunities in various medical domains where promising AI techniques, particularly Reinforcement learning (RL) based solutions can be applied. ...
Full-text available
A global health emergency has been declared by WHO at the beginning of 2020 based on the increasing number of cases in the COVID-19 epidemic. Governments around the world have taken unprecedented measures. However, there is no guarantee that the measures taken are best to mitigate the effect of pandemic. We investigate the impact of government policies regarding interventions on deaths related to the COVID-19 and mitigation of the economic decline. In a simulation environment, we use Reinforcement Learning (RL) to explore the optimal policies to prevent COVID-19 outbreak. We use a specific simulator called PandemicSimulator which has detailed abilities to simulate spread of disease and people interactions at different locations. The simulator is utilized to train RL agents to take mitigation policies with minimum economic damage of the pandemic without exceeding the hospital capacity. We use Deep Q Networks to train the RL agent. We compare the performance of the our agent’s policy with the policy applied by the United Kingdom in terms of critical patients, deaths and economic damage. Results show that policies improved by the RL agent can help decision makers in the pandemic mitigation policies.
ResearchGate has not been able to resolve any references for this publication.