The Knowledge Discovery in Healthcare Data (KDH) workshop series was established in 2016 to present AI research efforts to solve pressing problems in healthcare. Since, the successful series continued in 2017, 2018, 2019 and 2020 to bring together clinical and AI researchers to foster collaboration. This year, the workshop will be held in Captetown, South Africa, in conjunction with IJCAI 2023.

In alignment with this year IJCAI's theme of AI for Good, the focus of the workshop is on robust, flexible, generlisable, explainable and ethical applications of AI in medicine. The aim is to translate routinely collected clinical data into knowledge that drives the continual improvement of medical care and trustable applications that can improve the efficiency and productivity of healthcare professionals. The grand aim requires 1) the extraction, organisation and assembly of large amounts structured and free-text data embedded with electronic patient records,  2) the design of knowledge discovery and decision support tools that capitalise on the abundance of medical knowledge and guidelines in addition to the the large, temporal and uncertainty-ridden healthcare data and 3) the design of trustworthy models that are robust, and capable of reasoning about data biases and existing guidelines and workflows. Such requirements will lead to the ability to provide personalised recommendations and decision support tools to aid both patients and care providers, to improve outcomes and provide personalise care. Our scope extends to the range of applications initiating response to patient monitoring, for example, the prediction of cardiac arrest from sparse hospital wards data, and alerting patients or automatically adjusting insulin doses when blood glucose levels are predicted to go out of range.

Proceedings

Proceedings will be published via CEUR onlie proceedings and indexed by DBLP. The committee is currently setting up dissemination routes for a journal special issue. Confirmation of a special issue will be determined by the number and quality of the workshop submissions.

Submission Guidelines

All papers must be original and not simultaneously submitted to another IJCAI workshop. The following paper categories are welcome:

Submissions can be made as:

1. Long papers (6 pages + 1 page references): Long papers should present original research work and be no longer than seven pages in total: six pages for the main text of the paper (including all figures but excluding references), and one additional page for references.
2. Short papers (3 pages + 1 page references): Short papers may report on works in progress, descriptions of available datasets, as well as data collection efforts. Position papers regarding potential research challenges are also welcomed. Short paper submissions should be no longer than four pages in total: three pages for the main text of the paper (including all figures but excluding references), and one additional page for references.

Both long and short papers must be formatted according to IJCAI guidelines and submitted electronically through easychair.

List of Topics

• Machine Learning, Knowledge Discovery and Data Mining 
• Explainable, ethical and robust AI in medicine
• Representation and Reasoning about Medical Knowledge
• Medical Natural Language Processing
• Integration and use of Biomedical Ontologies and Terminologies   
• Medical applications of Autonomous and Multiagent Systems
• Biomedical Imaging and Signal Processing
• Advanced Biomedical analytics
• Privacy, security and reproducibility of Biomedical AI

Organizing Committee

• Dr Zina Ibrahim, Associate Professor in Artificial Intelligence in Medicine, King's College London, UK
• Dr Honghan Wu, Associate Professor in Health Informatics, University College London, UK
• Prof Nirmalie Wiratunga, Professor in Intelligence Systems, Robert Gordon University, UK

Publication

KDH-2023 proceedings will be published by CEUR (https://ceur-ws.org/).

Contact

All questions about submissions should be emailed to zina.ibrahim@kcl.ac.uk or honghan.wu@ucl.ac.uk