Boosting Health AI capabilities to better patient outcomes Page Content Healthcare is one of the five sectors in which the government has committed to boosting Artificial Intelligence (AI) capabilities under the National Artificial Intelligence Strategy (NAIS) that was launched by Deputy Prime Minister Heng Swee Keat on 13 December 2019.The NAIS supports projects in key high-value sectors to build a holistic AI ecosystem in Singapore, increase productivity and raise the quality of public goods and services. For healthcare, the aim is to alleviate one of Singapore’s pressing problems – the rising prevalence of chronic diseases and its related risk of complications such as a heart attack or a stroke. The SolutionAI can be used to slow the progression or prevent the development of chronic disease complications such as high blood pressure, high cholesterol and diabetes. Changes in the retinal blood vessels can provide useful information on an individual’s cardiovascular status. For example, narrowed vessels with thickened walls could be due to uncontrolled high blood pressure. Thus, the analysis of retinal images, and other clinical and biomarkers, can be used to create a predictive risk assessment model.This will be developed by building on the capabilities of the Singapore Eye LEsioN Analyzer (SELENA+), an AI-powered image reader in use today. AI can also be used to create a personalised care plan for patients to better support primary care teams. It can also empower them to take charge of their health by helping to nudge or remind patients to make lifestyle changes and improve medication adherence. This could be in the form of a mobile application or a wearable device. Overcoming ChallengesAI is still an emerging technology with complexities such as the ethics and governance of AI in health. This includes the use of data and its related security concerns. MOH and IHiS are working together to create a framework that is specific to healthcare. This will supplement the Personal Data Protection Commission’s Model AI Governance Framework announced early this year. Work to integrate backend systems and workflow processes for AI solutions such as the deployment of SELENA+ at designated polyclinics has also begun. These are but the first steps of a ten year journey to better utilise AI and transform care delivery at the primary care level. The OutcomeWhen the NAIS announcement was made, healthcare’s use case was widely covered by local and international media outlets. The use of AI in healthcare is an area in which major technology firms and governments across the globe are looking towards to reduce healthcare spending and improve patient outcomes, against a backdrop of rising chronic diseases and growing manpower constraints. If executed well, Singapore can become a leader in health AI. What’s Next?Over the next two years, SELENA+ will be deployed nationwide. Patients will be able to receive their results in minutes as compared to the current wait time of about a day. The use of the AI system also has the potential to reduce the workload of the human graders by up to 70 per cent.The sector will continue to work on creating impactful solutions. This means there is a need to go beyond the public healthcare sector and Institutes of Higher Learning to tap expertise from the private sector. The approach will enable the adoption of best practices and the sharing of solutions. To that end, IHiS will roll out two initiatives:(i) A Call-For-Collaboration (CFC) in February 2020 to invite interested companies to collaborate on the development of an AI-Enabled Medical Imaging Platform. The platform will allow various commercial and research AI partners to develop novel AI applications in healthcare based on the analysis of medical images. (ii) Grant call for AI medical imaging projects; IHiS will also support projects in obtaining regulatory approval and deployment plans. More details will be released in early 2020.If you’re interested to find out more on the CFC, please contact Kelvin Sng from IHiS’ Emerging Services and Capabilities Team at email@example.com.