ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2019) 63 NSA4 | DOI: 10.1530/endoabs.63.NSA4

How can AI create value for the endocrinologist? Challenges and opportunities for AI-enabled solutions

Anca Bucur


The Netherlands.


The risk of unsustainability faced by many healthcare systems drives the focus on leveraging technology for effective solutions to key problems. Technologies such as AI, enabling to better leverage the power of data in healthcare, hold the potential to play a transformative role and contribute to improvements in both efficiency and quality. AI-enabled solutions may lead to lower costs, better patient outcomes, increased safety, streamlined information flows, and free healthcare professionals of tedious and frustrating data handling tasks. These solutions must support patient-empowerment as well, turning patients into informed partners in the medical decision making processes and fostering a shift towards preventive care. An important enabler for AI is increasing data quality. Tailored data transformation and information extraction need to be developed to augment data quality at clinical sites and help leverage all relevant information. Techniques for at-integration data extraction and in-flight classification with AI models need to be combined to unlock valuable information. Emerging Open Data initiatives could add significant value by accelerating research and development in AI, reducing innovation cost, and introducing community-wide standards and best practices. The adoption of AI in healthcare relies on the ability to provide proper, privacy-preserving use of data, and to ensure that the outputs are adequately validated, transparent, and explainable. Moreover, the new solutions must seamlessly fit in the clinical workflows, which is a non-trivial challenge, and users need to be appropriately trained to use them and to correctly interpret the outputs. The diversity of business models under which different AI assets are monetized to make the whole ecosystem sustainable, complicates things even further. A healthcare-specific AI platform that helps efficiently address the above challenges is essential to scaling the development and deployment of AI assets. The talk will introduce our ongoing work and discuss new ideas for research and collaboration.

Article tools

My recent searches

No recent searches.