Ashkan Khakzar

Papers

Total Citations

H-Index

About

Ashkan Khakzar is a researcher whose work bridges the critical intersection of medical imaging, machine learning, and explainable artificial intelligence (XAI). His primary research areas include the reproducibility of ultrasound imaging, the interpretability of deep neural networks in medical contexts, and the development of robust, transparent AI systems for clinical decision support. Khakzar’s major contributions are twofold: first, he has advanced our understanding of how robotic and expert-operated ultrasound acquisitions compare in terms of reproducibility, a foundational issue for automating diagnostic imaging. Second, he has pioneered methods to explain the inner workings of convolutional neural networks (CNNs) in medical image analysis, helping to build trust in AI-driven diagnostics. His work on ultrasound reproducibility, with 73 citations, has been influential in shaping protocols for robotic-assisted procedures and has informed the design of more reliable imaging systems. Beyond this, his research on XAI has been recognized for its clarity and practical utility, often cited in studies aiming to demystify AI predictions for clinicians. Khakzar’s contributions are vital for ensuring that AI tools are not only accurate but also interpretable, a key step toward their safe integration into healthcare.