The Statistical Methods for Health Equity Series is a monthly online series co-hosted by the Data Science for Health Equity community, the Alan Turing Institute Health Equity Interest Group, and the Department of Statistical Science at University College London.

---

We are excited to announce the next instalment of our webinar series with speaker Nathan Blake from UCL and KCL.

Specific details on the topic are as below:

Topic: 

Explainable AI for clinical applications: Causal Responsibility-based Explanations

Abstract:

The deep learning paradigm holds much promise for clinical utility but faces a number of barriers. One of these is the inscrutable nature of AI models: even if performing well on traditional performance measures, there remains the question as to why a model reaches the decision that it does. Explainable AI (XAI) models seek to shine a light on these opaque models. In this talk we will explore the need for XAI in the clinical setting and the need for rigorously evaluating the explanations themselves. This will be done using a particular XAI tool, ReX, which has been developed within the framework of actual causality and counterfactual reasoning and has been applied to several medical data types.

Speaker Biography:

Nathan has 12 years of experience as a nurse, predominantly in A&E, before completing a PhD. in computational biology. His current work looks at applying causal reasoning for explaining the outputs of black-box AI models as applied to clinical problems. This involves numerous collaborations with clinicians including multiple data types including neuro-imaging via MRI, CT and PET, triaxial accelerometer cancer patient data to detect and explain early deterioration and vibrational spectroscopy to augment histopathological decisions in cancer prognosis. You can follow Nathan at his blog: https://nathanblake.co.uk/

--

Please direct any questions about the webinar series to Dr Brieuc Lehmann at b.lehmann@ucl.ac.uk.