Dr Liew using CAPE, an AI predictive tool.
When Dr Charlene Liew began practising radiology at CGH in 2009, she had opportunities to represent CGH at conferences locally and abroad, both as an educator and researcher. The interactions she enjoyed with top academics at these meetings broadened her perspective and ignited a spark within her to embark on her own innovation and research work at CGH.
The support and funding of CGH in Dr Liew’s pursuit of innovation projects enabled her to continue to develop partnerships and seize more opportunities. Soon after, Dr Liew co-founded the AI and Informatics subsection of the Singapore Radiological Society, which facilitates the development of artificial intelligence (AI) in radiology for the benefit of patients and radiologists. “My greatest honour to date, is to have been elected by my colleagues as the Chapter Chairperson for Diagnostic Radiology of the College of Radiologists, Singapore,” shares Dr Liew. “As the professional body in charge of advancing the art and science of medicine locally, it is a heavy responsibility to carry on the good work of those who built up the practice of Radiology.”
AI as a predictive tool
Dr Liew is part of the CGH team that developed an AI predictive tool that determines the likelihood of whether a patient has mild or severe pneumonia, based on chest X-ray images. Together with the Integrated Health Information System (IHiS) team, they developed the Community Acquired Pneumonia and COVID-19 Artificial Intelligence Predictive Engine (CAPE) within four months.
Being rapidly deployable, the AI tool enables the innovation and development of new responses against the virus, such as the development of vaccines and running of diagnostic tests in a short span of time. CAPE is also able to predict the likelihood of patients requiring critical care almost instantaneously, alerting medical teams of possible clinical deterioration and prescribing the appropriate interim measures to improve patient outcomes.
Beyond local healthcare settings, CAPE can also potentially be calibrated to identify and predict the severity of respiratory infections globally. This would be crucial during pandemics where there is an increased need for inpatient and critical care support. In areas where healthcare resources may be limited, CAPE can enable prioritisation of healthcare resources so that patients who are likely to develop severe pneumonia can receive appropriate and timely care, improving outcomes.
AI and digitisation in medical applications
Digital automation can help relieve doctors from the burden of manual entry of clinical records. Dr Liew highlights that the adoption of electronic medical records (EMR) has been a gradual, ongoing process over the past 10 years. EMR has been a great boon for healthcare systems, providing doctors with access to medical records from all interlinked public healthcare institutions, enabling seamless care for patients, who can then experience smooth transitions from the hospital to community, and from specialist centres to non-specialist centres.
Inevitably, the increasing digitisation of medical records will come with its challenges. Doctors spend more time on digital record entry, which can detract from time spent communicating with patients, although AI-assisted scribe systems which can accurately take down physician notes using voice commands may help alleviate this issue. Furthermore, EMR generates a huge amount of mostly unstructured data. The present challenge is to implement EMR systems which are contextual, searchable and structured, so that this data can be harnessed for population-based research leading to new insights and discovery.
“Most of the changes we experience will increase convenience for patients and caregivers,” says Dr Liew. “Some are already in use, such as tele-medicine for general practitioner (GP) consultations, the booking of appointments on your smartphone, and accessing your medical records online.” These help people to spend less time in queues, or shuttling about in hospitals, and make the running of people’s lives increasingly efficient. “Big data is poised to be the epidemiologist’s and public healthcare doctor’s silver bullet, allowing them to predict where the next cluster of flu or dengue fever will erupt so mitigating action can be taken.”
Dr Liew believes that the next wave of change will be in the sphere of clinical medicine. “These will include early warning systems for discharged patients at risk of readmission, inpatients at risk of acute injuries or life-threatening reactions to infections such as sepsis, and those at risk of falls,” she explains. These systems will positively impact lives, saving thousands and millions of quality-adjusted-life-years worldwide.
Eventually, digital transformation and artificial intelligence will transform the whole of healthcare and bring in new models of care, some of which are currently being developed. In the realm of imaging, cardiovascular risk may be detected through routine images taken for other purposes. Screening for cancer and neurodegenerative conditions in our bodies can be partially automated and may be as commonplace as sending our cars for an annual inspection. “We may be told at a young age which diets to adhere to and avoid; and at a ripe old age, which symptoms to be aware of and what personalised medicines to take,” says Dr Liew. “In this picture of the future, we should be prepared to live much longer, healthier lives.”
Medicine has always been, and will continue to be, a ‘high-touch’ profession. “Digital automation will allow doctors to be at our patients’ bedsides where we are needed most,” says Dr Liew. “Eventually, digitisation and AI will transform healthcare and introduce new models of care.”
“My chosen specialty, Radiology, is right at the forefront of this technological change, with more than 70 per cent of our work taking place in the digital space. Much like other industries which have undergone disruption, we were the proverbial canary in the coalmine — among the first to be impacted significantly.”
Subscribe to our mailing list to get the updates to your email inbox...