G-Agent in Healthcare

Given that healthcare services are complex, a pure delegation of authority without proper governance is not good enough. The example thus shows how Diagnostics AI can be matched with a G-Agent based governance:

A digitalization strategy for healthcare providers allows to holistically answer the question how digitalization looks like. Such overall strategic view starts from key trends in Healthcare or patient needs and includes benefits, required processes, architecture as well as digital skills. A strategy-led approach shall address digitalization beyond use-cases-only and, therefore, not “miss the wood for the trees”. Healthcare providers must be leveraging on proven new technology like AI, VR, ML, Cloud, Robotics, etc., systematically.

In this context, AI can ensure both leveraging current expertise of experienced practitioners across the organization as well as further developing this expertise by leveraging data and creating evidence for even better therapies amongst others. The compelling use of AI and data will thus help to explore and improve effectiveness beyond conventional QA. The AI driven approach will include both comprehensive data aggregation across sources and treatment area specific deep-dive analytics.

To make comprehensive use of such Diagnostics AI, data governance must evolve towards AI governance. Scientific experiences about developing evidence through real-world data, instantly (i.e., over and beyond ex-ante clinical trials which provide real-world evidence) have been rather pre-mature so far. The G-Agent helps to design the process thoroughly to maintain the Chief Medical Officers` comfort feeling in hospitals in the future and leverage AI, boldly. AI systems are supposed to refer to real-world-evidence for diagnostics and treatments. Since this must be clearly defined within the organization, i.e. as roles and responsibilities, the authority of such AI must be included. The G-Agent can help to monitor compliance of such AI. Newly created evidence based on real-world data must be clearly distinguished and is subject to approvals. The G-Agent, therefore, is asking for a real change process – not limited to a “technical” governance topic – where tech and med cultures come together, appreciate each other and create a new, innovation-driven Digital Health Care mindset.

Let’s take the example of representing real-world evidence related to breast cancer detection. This can be complex, but here is a simplified scenario to illustrate the subject:

a) Data source: Real-world evidence: Databases with patient data, diagnostics and treatment histories.

b) G-Agent Functions (including GAuth+):
Data validation: The G-Agent receives data from validated medical databases.
Condition check: The G-Agent checks whether the data meets certain criteria (e.g. age, family history, mammography results).
Diagnostic validation: Based on the predefined medical criteria, the G-Agent provides a diagnostic validation, which is being compared with the diagnostics of the AI system authorized by the doctor. The G-Agent creates an alert in case of deviations.

c) Gimel Governance process:
Expert review: The healthcare provider or doctor reviews the G-Agent alerts, compares it with the AI`s diagnostics and either confirms or rejects it.

Automated notification: The G-Agent automatically sends notifications to patients or doctors based on the validation results. If new insights based on real-data evidence is being created, reviews and approvals will be initiated to fuel both the AI as well as G-Agent with new intelligence.

This is a simplified example, but in practice, additional functions and security measures would be considered to create a fully functional and secure AI governance within the Gimel operating system. If you would like to dive deeper or have any questions, we are happy to help!