Keeping a close AI on the future

By Kami Navarro

For people in Singapore, the long-awaited new normal is finally upon us. After nearly two months of coronavirus restrictions, the nation is beginning to cautiously reopen the economy. Dining and retail outlets are buzzing with activity again, and talks are already in place to restart international travel.

Despite these welcome developments, it’s important to remember that the COVID-19 pandemic is far from over. Until an effective vaccine becomes widely available, resuming economic and social activities will always inherently carry health risks. Decision-makers across the public and private sector now have to grapple with the question: How can these health risks be effectively managed?

The answer lies in artificial intelligence (AI), according to Mr Neal Liu, co-founder and CTO of predictive AI company UCARE.AI in Singapore. In 2016, Mr Liu and his team of data scientists and technologists founded UCARE.AI to fulfil one mission: to use data ethically to solve real-world problems and improve lives. To achieve the highest possible social impact, the UCARE.AI is starting with AI applications in health. 

Stepping up screening

COVID-19 cases have so far been confirmed in nearly 200 countries and territories. In the early days of the outbreak, cases were largely seeded by international travellers—resulting in the travel industry grinding to a virtual halt. As countries slowly reopen their borders, stringent measures are needed to prevent imported cases.

Applying their expertise in AI to the situation, UCARE.AI and their collaborators developed a turnkey solution dubbed TravelTech to supplement thermal screening at border crossings. Once deployed, the TravelTech solution is set to screen and classify travellers according to their risk of having COVID-19. The resulting data could then be used to prioritise testing and reduce instances of high-risk passengers slipping under the radar.

Though the TravelTech solution is still in its prototype phase, having such measures in place will surely help revitalise the travel sector soon enough.

Predicting patient payments

“Our aim is to apply our predictive engine to all aspects of the healthcare industry by creating the most advanced AI capable of making accurate predictions years into the future,” revealed Mr Liu. Partnering with Parkway Pantai, UCARE.AI successfully launched in 2018 the AI-Powered Pre-Admission Cost of Hospitalisation Estimation (APACHE™) at Parkway’s various locations across Singapore.

Typically, hospitals estimate the final bills of patients based on historical bill sizes from previous admissions. However, this method fails to account for unexpected events like complications that result in a longer stay or additional surgeries. Therefore, by providing more accurate estimates, Parkway hoped to enhance the transparency of their hospital charges and build patient trust.

This led UCARE.AI to develop APACHE^TM, an AI model that estimates bill sizes based on more contextual factors including the patient’s medical condition and visit frequency, diagnosis, surgical procedures performed, and price fluctuations.

The model has since been successfully deployed in all Parkway hospitals.  

For their groundbreaking solution, UCARE.AI was cited by the Infocomm Media Development Authority of Singapore (IMDA) and Personal Data Protection Commission in their Model AI Governance Framework as a prime example of a large-scale AI use case. At the present, UCARE.AI is in talks to deploy the model to other hospitals in the Asia-Pacific region.

Going beyond health

With their success in the healthcare industry, UCARE.AI is now looking to apply their machine learning algorithms and cloud-based AI platform known as AlgoBox to other sectors, including insurance, finance, travel, and government. Regardless of sector, their AI machine learning algorithms could be used to predict costs and customer profiles, forecast demand, and even detect anomalies in data. AlgoBox, on the other hand, is meant to empower clients by allowing them to easily train, develop, and deploy new AI models. 

As a testament to the strength of their products, UCARE.AI was accredited by IMDAunder their Accreditation@SGD programme. “Being accredited by IMDA has put us on the fast track for global growth by providing an assurance of our product core functionalities and ability to deliver, as well as establishing UCARE.AI’s credentials,” said Mr Liu. IMDA has also provided vital support in opening doors to potential clients through features in media and events, he added.

For a company founded only four years ago, UCARE.AI has quickly amassed an impressive set of accomplishments. Even without the help of predictive AI, it’s crystal clear that UCARE.AI is set to make its mark in healthcare and other industries.