News: New Centre for Biomedical Informatics to transform healthcare with data analytics and AI

By Sarah Zulkifli, Science Writer, Communications & Outreach

On 21 April, LKCMedicine launched the Centre for Biomedical Informatics (CBI) that will leverage data analytics and artificial intelligence (AI) to develop “super algorithms” that predict and personalise treatment in areas such as mental health.

From left: CBI Advisor and NUS Department of Biochemistry A/Prof Tan Tin Wee, LKCMedicine Dean Prof Joseph Sung, LKCMedicine Vice-Dean (Research) Prof Lim Kah Leong, and CBI Co-director Asst Prof Bernett Lee

The Centre will use its expertise and state-of-the-art equipment to identify trends, patterns, and anomalies in data to derive insights that will help researchers and clinicians make better informed decisions, and possibly give rise to new discoveries and the development of powerful diagnostic and treatment methods for diseases.

The work at the Centre is in line with the National AI Strategy under Singapore’s Smart Nation initiatives to deepen the use of AI to transform Singapore’s economy. One key area outlined in the strategy is in healthcare, where chronic disease prediction and management could help with faster detection and treatment of such diseases. 

Led by Co-directors Assistant Professors Bernett Lee and Wilson Goh, the Centre will leverage on the unique strengths of NTU and LKCMedicine, and focus on collaborating with world-class academics in engineering, humanities, education, and business to advance medicine.

CBI Co-directors Asst Prof Bernett Lee (left) and Asst Prof Wilson Goh

As part of the Data Science cross-cutting research theme at LKCMedicine, CBI will serve as an interface to the wider research community bridging the gap between medical professionals and researchers from engineering, social sciences, and business. The Centre will also play a role in educating future medical professionals and researchers in the understanding of how data contributes to impactful discoveries. 

Celebrating the Centre’s launch

Held at the Experimental Medicine Building Learning Studio on NTU Singapore’s main campus, the launch was attended by the LKCMedicine community as well as healthcare leaders and professionals from Tan Tock Seng Hospital, National Supercomputing Centre Singapore and A*STAR.

In his opening speech, LKCMedicine Dean Professor Joseph Sung said, “The opening of the Centre is very timely; it will not just support LKCMedicine’s five flagship research programmes, but also research by our partners in NTU and beyond. We need good quality data so that we can make important advancements in science.”

LKCMedicine Vice-Dean (Research) Professor Lim Kah Leong, who spoke next, added, “The establishment of the Centre will help LKCMedicine to define and establish a core in biomedical informatics. Data is the new oil in research and medicine. Like oil, data is worthless in its raw form. It requires refinement, cleaning, structuring and amalgamation. Also, like oil, data has a multiplicity of end uses for instance in developing algorithms to detect diseases from images, formulating approaches to help busy hospitals respond to emerging needs, and deriving insights from randomised registry trials.” 

Speaking on creating close working ties with the National Supercomputing Centre Singapore, CBI Advisor and Associate Professor of National University of Singapore Department of Biochemistry Tan Tin Wee said, “The next vision is to become the national quantum computing facility that will create a new paradigm of solving problems. We now have the ability to provide computing power for biomedical research. And we are not stopping there. Plans are underway to build a data centre in NTU itself. This is my mission of petascale computing. We are looking at how Singapore can move towards pre-exascale and hopefully exascale as well.” 

Asst Prof Lee went on to share CBI’s vision and mission. “To build up biomedical informatics capabilities among medical science researchers in Singapore and the region, the Centre will also organise workshops and courses. Through these efforts, we hope to bridge the gap between medicine and data, devise meaningful and practical ways to confront the complexities in biomedical data head-on, and bring forth a new wave of personalised models and therapies for predictive and preventive disease management,” he added.

Next, Asst Prof Goh gave an overview of the academic and research vision and alliances for the data science research cluster. He said, “At CBI, we are making sense of huge volumes of biological data. We want to work towards achieving the three ‘P’s in clinical application: prediction, prevention, and personalisation. By building biologically informed models through data analysis and super algorithms, we could create insights that are personalised for the patient. Such models could enable early and accurate detection and prevention of chronic diseases and acute medical emergencies.” 

Prof Sung, Prof Lim, A/Prof Tan and Asst Prof Lee were then invited on stage to mark the opening of the Centre by pushing a button. This was followed by a tour of the facilities and lunch for the invited guests.

Research projects on mental health conditions and cancer

Among the Centre’s projects is an ongoing collaboration with the Institute of Mental Health (IMH) and the Auckland University of Technology (AUT) to better understand and predict disease progression of mental health conditions in youths using data analytics and AI techniques. The team is conducting a four-year study to develop machine learning methods using a combination of different datasets to enhance the accuracy of early detection or prediction of mental illnesses in at-risk youths. 

This collaboration taps into data from IMH’s 2009 longitudinal youth-at-risk study that studied 600 youths to identify social, biological, clinical, and cognitive factors involved in the transition to psychosis in at-risk youths.

Applying advanced neural networks (a machine learning technique), the team will analyse and integrate a wide variety of data, including clinical, behavioural, and large-scale molecular data to understand how seemingly disparate data relate and connect to one another. Such an approach opens doors to discoveries of new biomarkers and risk factors for the screening of mental health states.

These new discoveries could then be used to develop super algorithms that could one day predict who is at risk of mental disorders, allowing for better clinical intervention via early prognosis and diagnosis of mental health issues in at-risk youth.

The findings could also help in the development of personalised modelling for a better understanding of individual factors that trigger mental illnesses.

Aside from mental health, the Centre is also working on projects in the area of cancer treatment. For instance, using statistical, meta-analysis, and machine learning techniques, researchers have devised a way to develop cancer biomarkers – biological molecules that are a sign of disease. They have used this method to produce a breast cancer biomarker associated with prognosis.

Based on this biomarker, the researchers are now developing novel therapeutic strategies to help them discover drugs capable of “reversing” the biomarker expression patterns in hopes of improving patient outcomes.