Presentation at CRI’2025 Conference

CRI'2025 - The 7th edition of Conference on Research in Computer Science CRI’2025

Location: Yaoundé, Cameroon
Date: 16 Decembre 2025
Event: The seventh edition of Conference on Research in Computer Science –Conférence de Recherche en Informatique– CRI’2025 (CRI'2025)

Presentation Details

At CRI'2025, I presented my research on "Still Malignant if You Were a Woman? Auditing Group and Counterfactual Fairness in Dermatological AI." This work addresses critical fairness concerns in AI-powered dermatological diagnosis systems, examining how these models may exhibit biases across different demographic groups.

The presentation showcased a comprehensive fairness auditing framework that combines group fairness metrics to assess disparate performance across demographic groups, counterfactual fairness analysis to evaluate whether predictions would change if sensitive attributes were different, and clinical validation using dermatological datasets to identify potential biases in skin lesion classification. This research highlights the urgent need for fairness-aware AI in healthcare applications, particularly in dermatology where existing datasets and models may underrepresent certain skin tones and demographic groups.

Abstract & Publication

Status: Accepted at CRI'2025
Type: Full paper presentation

The paper proposed multimodal CNN and provided an audit of fairness disparities in dermatological AI(skin lesion) systems and proposes methodologies for auditing to ensure equitable healthcare delivery.

Conference Impact

CRI'2025 brings together researchers, academics, and industry professionals to discuss advances in computer science, artificial intelligence, and their applications across various domains. The conference emphasizes the importance of responsible AI development and ethical considerations in deploying machine learning systems, particularly in high-stakes domains like healthcare.

Key Themes

• Fairness in AI and machine learning
• Healthcare AI and medical imaging
• Counterfactual fairness analysis
• Responsible AI development

Related Publications

For more information you can follow this link:
Still Malignant if You Were a Woman? Auditing Group and Counterfactual Fairness in Dermatological AI