The 6th international workshop on 

TOPICS & SCOPE

Contributions are welcome in in all aspects of the use of Artificial Intelligence in medicine and healthcare, including but not limited to the following:

• Representation, Reasoning and Learning from Medical Knowledge

  1. Temporal data and knowledge representation in biomedicine, 

  2. Representation of uncertainty, including probabilistic and fuzzy representations of biomedical data.

  3. Integration and use of Biomedical Ontologies and Terminologies   

  4. Multilevel data integration in healthcare, e.g. behavioural data, diagnoses, vitals, radiology imaging, Doctor's notes and different omics data, including multi-agent approaches.

  5. Methodologies for developing hybrid medical AI systems that can achieve robust and scalable prognosis and prediction by capturing the ability to represent and reasoning about rich existing domain knowledge, e.g. clinical guidelines, to steer and meaningfully interpret a Machine Learning model output, i.e. neurosymbolic systems. 

  6. Methodologies for embedding medical knowledge, e.g. clinical guidelines, into Machine Learning and other approaches to solve healthcare problems. 

• Machine Learning and Data Mining for biomedical analytics and knowledge discovery 

  1. Methodological contributions using machine learning to solve healthcare problems on all scales, from biomedical knowledge discovery to clinical outcome prediction.

  2. Examination, reflection and solutions to challenges and issues related to devising machine learning solutions to biomedical problems, such as incorporating domain and user preferences  interpretability, algorithmic robustness, bias and multimodality. 

  3. The availability and quality of data, including the quantification of data dimensionlity, learning under sparse settings, data standardisation and efforts to create open data repositories. 

• Medical Natural Language Processing

  1. Information retrieval, entity recognition and document classification 

  2. Knowledge abstraction, classification, and summarization from literature or electronic health records

• Privacy and Ethics (High quality submissions in this area will be recommended to a BMC collection https://www.biomedcentral.com/collections/eaihm.)

  1. Bias and inequality associated with health data and Machine Learning in medicine

  2. Elaboration of novel ethics-relevant metrics

  3. Research on attitudes and perceptions of physicians and the public on the implementation of Machine Learning and other AI solutions to healthcare problems. 

  4. Ethics surrounding AI-associated privacy and surveillance

  5. Ethical challenges and solutions surrounding the implementation of Machine Learning in medicine. 

• Scale, Deployment and Reproducibility 

  1. The reproducibility crisis in Machine Learning and its effects on the successful implementation of Machine learning innovations in biomedicine. Approaches to embed reproducibility in biomedical MLOps, including model versioning, automation and federated training at scale. 

  2. Open source technologies for MLOps and their use in biomedicine. 

  3. Experiment, data tracking, and training at scale. 

  4. Machine learning deployment and operation challenges, including deployment and monitoring in production. 

• Advanced biomedical data processing

  1. Imputation and dealing with non-uniform sampling of patient and other healthcare data, dimensionality recognition and feature engineering

  2. Issues relating to data heterogeneity and scale

  3. The identification of sources and forms of data bias, as well as methodological solutions. 

  4. Visual Analytics in Biomedicine

  5. Models for handling uncertainty in biomedical data

  6. The use and processing of multimodal data including imaging, speech and other signals.