As biomedical research moves towards the era of big data, massive amounts of complex data are being generated. Artificial Intelligence (AI) and machine learning (ML) technologies have been increasingly important to fully extract value from rich and complex datasets to drive scientific discoveries and clinical decision making. However, the adoption of AI and ML in endocrinology and metabolic diseases is lagging behind, compared to fields such as cancer genomics. A major part of the challenge comes from the complexity and heterogeneity of data being produced by different platforms and research groups, the lack of data standards, data sharing platforms, and data processing pipeline widely accepted by the community. To fully leverage the power of AI to maximize the value of the rich data in endocrinology and metabolism, at least a few key areas need to be addressed. First, aggregated, harmonized, discoverable, and accessible datasets that are suitable for ML and AI applications are in urgent need. To this end, developing data sharing infrastructure, standards, and data curation pipelines are the key to success. Second, ML and AI tools and algorithms that automate or semi-automate the process of data curation, harmonization, annotation will accelerate this process. In this context, multi-omic data integration methods based on ML and modeling approaches can combine different types of data and improve the mechanistic characterization of diseases. Third, it is crucial to develop easy-to-use tools and visualization methods that can be used by researchers and clinicians not trained as computer scientists to drive scientific discoveries and clinical decisions.

We invite submissions that address applications of ML and AI to challenges related to the discovery, reuse, curation, management, and sharing of endocrinology and metabolism data.
Topics of interest include novel algorithms, tools, platforms, datasets, and successful use cases. All article types accepted at Frontiers in Endocrinology will be considered.

Specific topics include but are not limited to:

● Efforts building data sharing portals, knowledge bases, and data discovery platforms
● Generation of machine readable datasets suitable for ML and AI applications
● Automated or human-in-the-loop methods for data curation, annotation, integration, and metadata extraction
● Automated or human-in-the-loop methods for measuring and improving data quality
● Tools that facilitate explainability of AI/ML outcomes and data visualization
● ML and AI applications for multi-omics and EHR data integration and analysis
● Algorithms, methods and tools for metabolic diseases diagnosis, monitoring, management, and outcome prediction
● Methods for multi-omic and multi-source data integration in endocrinology

Submit your idea

Impact

Frontiers in Endocrinology

Participating Journals

Impact