Eye tracking insights into physician behaviour with safe and unsafe explainable AI recommendations

M Nagendran, P Festor, M Komorowski… - NPJ Digital …, 2024 - nature.com
We studied clinical AI-supported decision-making as an example of a high-stakes setting in
which explainable AI (XAI) has been proposed as useful (by theoretically providing …

MIDAS: Deep learning human action intention prediction from natural eye movement patterns

P Festor, A Shafti, A Harston, M Li, P Orlov… - arXiv preprint arXiv …, 2022 - arxiv.org
Eye movements have long been studied as a window into the attentional mechanisms of the
human brain and made accessible as novelty style human-machine interfaces. However, not …

Eye-tracking of clinician behaviour with explainable AI decision support: a high-fidelity simulation study

M Nagendran, P Festor, M Komorowski… - ICML 3rd Workshop on … - openreview.net
Explainable AI (XAI) is seen as important for AI-driven clinical decision support tools but
most XAI has been evaluated on non-expert populations for proxy tasks and in low-fidelity …

Evaluating the impact of explainable RL on physician decision-making in high-fidelity simulations: insights from eye-tracking metrics

P Festor, M Nagendran, M Komorowski… - Workshop on Interpretable … - openreview.net
Explainable reinforcement learning (XRL) is crucial for reinforcement learning (RL)
algorithms within clinical decision support systems. However, most XRL evaluations have …

Interaction of doctors with explainable RL decision support via behavioural readouts of eye-tracking

M Nagendran, P Festor, M Komorowski… - … European Workshop on … - openreview.net
Explainable reinforcement learning (XRL) is crucial for reinforcement learning (RL)
algorithms within clinical decision support systems. However, most XRL evaluations have …

Clinical Evaluation Framework Using Behavioural & Visual Attention Read-Outs for Explainable AI (XAI)

M Nagendran, PGRE Festor, M Komorowski… - Available at SSRN … - papers.ssrn.com
Background: Explainable AI (XAI) is seen as important for clinical AI-driven recommender
systems but systematic evaluation of impact is lacking (typically only non-expert populations …