IJCNN
2021, Virtual event
Special Session on
Transparent and Explainable Artificial Intelligence (XAI)
for Health
Submission
link
Aims & Scope From the widespread
implementation and use of electronic health records to basic
research in pharma, and from the popularization of health
wearables to the digitalization of procedures at the point of
care, the domains of medicine and healthcare are bringing data to
the fore of their practice. The abundance of data in turn calls
for methods that allow transforming such raw information into
novel knowledge that is truly usable, including high stakes
decision support.
Machine Learning (ML) is enjoying unprecedented attention in
healthcare and medicine, riding the current wave of popularity of
deep learning (DL) and the umbrella concept of Big Data. But such
attention may bear little fruit unless data scientists effectively
address one major limitation that is particularly sensitive in the
medical domain: the lack of interpretability of many ML approaches
and, particularly, DL methods, leading in turn to limited
explainability. This may limit ML to niche applications and poses
a significant risk of costly mistakes without the mitigation of a
sound understanding of the flow of information in the model.
Domains where decision-making impacts our health motivate this
special session, to which we invite current research on
eXplainable Artificial Intelligence (XAI). The goal of XAI is the
design of techniques and approaches that still retain model
performance, while being able to explain their outputs in
human-understandable terms. With these capabilities, clinical
practitioners will be able to integrate the models into their own
reasoning, gaining insights about the data and checking
compatibility with working guidelines at the point-of-care.
This session aims to explore such performance-versus-explanation
trade-off space for medical and healthcare applications of ML. We
aim to bring together researchers from different fields to discuss
key issues related to the research and applications of XAI methods
and to share their experiences of solving problems in medicine and
healthcare. Applications leading towards routine clinical practice
are particularly welcome.
Topics that are of interest to this session include but are not
limited to:
• Interpretable ML Models in medicine and
healthcare: theoretical and practical developments
• XAI for electronic health records
• Integration of XAI in medical devices
• Human-in-the-loop ML: bridging the gap
between data and medical experts
• Interpretability through Data
Visualization
• Interpretable ML pipelines in medicine
and healthcare
• Query Interfaces for DL
• Active and Transfer learning
• Relevance and Metric Learning
• Deep Neural Reasoning
• Interfaces with Rule-Based Reasoning,
Fuzzy Logic and Natural Language Processing
• Assessment of bias and discrimination
in databased models
Preliminary Dates
Paper submission: January 15, 2021
Paper decision notification: March 15, 2021
Session Chairs
Alfredo Vellido
(avellido@cs.upc.edu), Universitat Politècnica de Catalunya,
Spain
Paulo
Lisboa (P.J.Lisboa@ljmu.ac.uk), Liverpool John Moores University,
U.K.
José D. Martín
(jose.d.martin@uv.es) Universitat de València, Spain