Artificial intelligence is rapidly evolving and has transformative potential in the healthcare sector. It is already making a difference in some areas; and we need to harness the emerging opportunities it presents, while maintaining safety and quality in healthcare.
Artificial intelligence (AI) includes systems, software, intelligent processes and tools, ranging in sophistication and complexity. While some applications – such as ChatGPT – are relatively new, others have been in use for some time.
In health, AI is already helping to automate repetitive processes, use data analysis to inform health service planning, and collect and transmit vital signs from wearable technology, such as smartwatches.
AI has the potential to enhance clinical decisions; not replace them. It can reduce the burden of manual data collection and administrative tasks; increasing the time clinicians can spend with patients.
Last year, the Agency for Clinical Innovation (ACI) supported an AI-based initiative using natural language processing to read free-text data in the NSW Health electronic medical record and electronic record for intensive care. Using this approach, we were able to identify patients who acquired a pressure injury in intensive care with 97% accuracy. This rapid process insight has the power to help inform clinical decision making in intensive care units (ICUs).
Dynamic simulation modelling is another example where we can create a virtual world on a computer to imitate how things change and interact over time in the real world. It helps us predict what might happen under different scenarios and test ideas in complex systems.
The ACI is now working with key partners to develop dynamic simulation models that inform service planning and delivery in clinical areas such as spinal cord injury, emergency departments and paediatric ICUs.
Emerging opportunities and global trends
Every day, we are seeing new and exciting AI tools emerge. It’s a busy space that changes quickly.
The NSW Health Critical Intelligence Unit (CIU) is helping us stay at the forefront of new and promising uses of AI in healthcare. A new, weekly Evidence Digest e-newsletter shares the latest information from international sources on clinical innovations that have the potential to change clinical practice and delivery, or organisation of care. It also includes the latest in AI health innovations.
The CIU also maintains living evidence on AI, which summarises new and promising ways AI can be used to support healthcare. Topics covered in the series include:
- clinical applications of AI
- automating indirect clinical tasks and administration
- system implementation.
As we maximise the potential benefits of AI-related innovations, it’s important we manage the risks. Maintaining privacy, safety and quality in healthcare is paramount. In the same way we approach any new treatment or clinical procedure, evaluating AI tools for clinical effectiveness and safety is crucial.
How do we know what's appropriate for NSW?
AI is increasingly being used in healthcare systems across the world. While we stay aware of emerging opportunities and seek advice from peak health and digital agencies, it’s important we apply a state lens to understand how we can use AI technology effectively in the context of NSW Health.
We must consider the practical steps and processes needed to ensure the quality and safety of AI applications in the NSW Health system, including: who will provide guidance on the safe use of AI, and how do we know if we are collecting the correct information and registering what we’re doing appropriately?
A dedicated NSW Health AI Taskforce has been established to help answer these questions and many more. It will inform and guide how AI is used in the public health system. Leveraging the knowledge and skills of around 25 subject matter experts and senior leaders from across NSW Health, the AI Taskforce will be supported by four working groups to develop an AI Framework.
I am pleased to lead this important work as co-chair of the AI Taskforce, with Dr Zoran Bolevich, Chief Executive, eHealth NSW, and Chief Information Officer, NSW Health.
The Framework will include important elements such as governance, communication, skills and capability, assets and intellectual property, data governance; and it will align with national and state government frameworks. It will inform our work going forward and help us understand the real-life application of these considerations for AI in healthcare.
Jamie Gabriel is the Chief Data Officer at the Cancer Institute NSW, and a member of the AI Taskforce. In our guest editorial, he provides an insightful perspective on the need to balance the potential of new technologies with trust, safety and strong governance.