What NUHS’ digital front door can tell us about the future of population health

By Amazon Web Services

A regional health cluster in Singapore is developing a conversational chatbot to improve patient outcomes and reduce hospital workload. Amazon Web Services delves into the artificial intelligence platform driving health innovation at NUHS to new heights and the cloud services accelerating healthcare services.

Singapore's National University Health System (NUHS) is implementing a conversational chatbot to better manage population health and support the tertiary care system. (Image: Canva)

When a patient is rushing to a clinic appointment in Singapore’s National University Hospital, she can now drop a message to the OneNUHS conversational chatbot to ask for directions.


The chatbot will not only be able to point her in the right direction, it will also be able to respond appropriately to her state of mind, help her request a medical report conveniently, and answer any questions she has. When she returns home, she can use the chatbot to ask further questions about her condition, and even initiate a privacy-preserving chat with her specialist if needed.


At the recent AWS Public Sector Summit 2022, A/Prof Ngiam Kee Yuan, Group Chief Technology Officer of National University Health System shared an insider look into the upcoming OneNUHS chatbot built on Amazon Lex, a service that can deploy conversational AI across all channels. 


Beyond providing patients with answers to queries regarding issues such as finance, navigation, and medical reports, it is expected that the chatbot will play a critical role in managing chronic conditions at a population scale for the National University Health System (NUHS), said Dr Ngiam. 


As part of Singapore’s new population health strategy, Healthier SG, Singapore’s health clusters will be responsible for managing the health of residents by regions, stated Minister for Health Ong Ye Kung in 2022. This means that NUHS will be responsible for overseeing the health of around 1.2 million Singaporeans in western Singapore, Dr Ngiam said.


NUHS’ digital front door 


The importance of managing population health was never more evident than during Covid-19. Observations made in clinics regarding individual patients flowed into policymaking, and were used to update processes and standard operating procedures, which were then communicated to patients. 


Covid patients were able to submit test results from the comfort of their home, allowing health agencies to monitor the spread of Covid-19 without overwhelming hospitals and clinics. When patients are able to manage their health at home, the healthcare system can manage more patients while lowering costs.


With OneNUHS’ chatbot, patients are able to submit their vital signs such as blood sugar levels and pulse rate for their care team to monitor. The care team can also choose to send nudges based on the readings. This means if a patient submits an abnormal reading, the care team can suggest the patient visit a doctor.


The chatbot is hosted on NUHS’ OneNUHS app, a mobile application that provides patients with telehealth services such as video consultations and services such as making appointments, paying bills, and requesting medical reports. The app uses AWS services like AWS Lambda and Amazon DynamoDB for application services, which allows the app to scale up and down to meet varying demand, wrote AWS.


While the patient communicates with the OneNUHS’ chatbot, a backend artificial intelligence platform, ENDEAVOUR AI, processes data continuously and helps care teams generate potential interventions, Dr Ngiam said to GovInsider. In 2022, The Straits Times also reported that ENDEAVOUR AI is being used to optimise bed capacity and manpower deployment.


The WhatsApp iteration of the chatbot is currently in its pilot phase, said Dr Ngiam. It was critical to host the chatbot on WhatsApp as NUHS’ team wanted to make the chatbot accessible for elderly citizens, who may find downloading and navigating an app a challenge. 


“Aspirationally, what we want to build is a conversational chatbot on AWS,” said Dr Ngiam in his presentation. This would allow the chatbot to respond empathetically to the feelings expressed by patients.


Continuous data for continuous care


But before such a digital front door is set up to improve patient engagement, agencies need to modernise their services through a cost-effective, modern data platform. 


First, data needs to be migrated to the cloud so that agencies can tap on new and upcoming technologies such as artificial intelligence and benefit from cost savings. Besides enabling the function of chatbots, this can enable agencies to drive further insights and move towards end goals such as improving population health strategies and precision health.


In the United Kingdom, National Health Service (NHS) Digital tapped on its AWS platform for building, scaling and deploying new products at a breakneck speed during Covid-19, shared Sarah Wilkinson, Chief Executive Officer at the 2021 AWS Public Sector Summit


Wilkinson highlighted that AWS HealthLake supported NHS in securely managing the data of more than 53 million individuals, building the digital infrastructure needed to support the extension of test facilities, and identifying the clinically vulnerable during the pandemic.


Massachusetts-based health network, Tufts Medicine (previously Wellforce), announced in 2021 that it would migrate its entire digital healthcare ecosystem to AWS. This would help them move from an “outdated, non-interoperable system” to one where data can be seamlessly integrated and analysed across applications, electronic health records, databases, and wearable technology seamlessly, wrote AWS. This will also enable it to tap on AI to improve patient outcomes.


Advancing precision medicine and genomics 


Finally, cloud technology can advance research and development capabilities in emerging fields such as precision healthcare and genomics. For instance, Genomics England turned to Lifebit CloudOS to securely compute large amounts of genomics datasets in seconds during the Covid-19 pandemic.


What made Lifebit stand out from older models of genomics research was its ability to bring its compute engine and analytics to the data itself, thanks to the Amazon Elastic Compute Cloud (EC2), wrote AWS. Its highly scalable cloud capabilities enabled Lifebit to work on datasets exceeding more than 100 petabytes of stored data. 


Cloud solutions such as Amazon Omics have helped other genomics organisations accelerate the translation of raw sequencing data into actionable insights quickly and securely, while being able to scale up quickly to match the large amounts of data genomics research entails.  


The secure, seamless exchange of data across healthcare databases and patient engagement platforms will be crucial for AI to improve health outcomes. In turn, this can pave the way for personalised healthcare and emerging healthcare opportunities in areas like genomics.


To learn more about how AWS cloud services is driving the future of health across artificial intelligence, population health, precision medicine, and genomics, download this eBook to uncover how some AWS customers are using the cloud to innovate and transform the way they work.