When AI meets healthcare
31 Aug 2021
Many years ago, a person's medical records comprised a few sheets of paper, such as photocopies and printouts of test results, in a folder. Over the past few decades, however, these simple folders have been transformed by the digitalisation process.
The concept of electronic health records (EHRs) was put forward in Hong Kong in 2008, with the aim of building up lifelong and more comprehensive digital health records for members of the public. Launched in 2016, the "eHealth" system in Hong Kong enables two-way sharing among public and private healthcare providers, which can facilitate the provision of more timely and accurate diagnosis and treatment for patients.
The implementation of EHRs also serves an even bigger purpose. According to Dr Yang Xian, Assistant Professor of the Department of Computer Science and an expert in natural language processing for EHRs, the digitalisation of health information has provided insights into the development of personalised healthcare.
Pointing the way towards personalised healthcare
Dr Yang has been investigating the application of artificial intelligence (AI), including the use of natural language processing in the capturing and interpretation of data from EHRs.
"The information extracted from the EHRs can be used to recommend drugs that will be effective for a particular patient," says Dr Yang. "By enabling healthcare providers to review the patient's medical history quickly, this can also aid the diagnosis of different diseases." She adds that a further benefit of EHRs is that they can be used to predict the risk of potential diseases, so timely interventions can be administered to the patient.
The adoption of AI technology in the medical field has paved the way for the implementation of personalised healthcare, which takes into account individual patients' unique clinical history and genetics to determine the best treatment. "Many diseases are not a single disease but a complex of different conditions. A personalised treatment regime can be tailored to the needs of each individual," Dr Yang says.
She adds, "To be able to deliver personalised healthcare, we need to leverage the data from various sources, including the EHRs, omics and clinical datasets. This is one of the potential directions in terms of the development of data-driven healthcare."
Merging healthcare and technology
Dr Yang's name in Chinese, Xian, refers to the chemical element Xenon, and this matches perfectly with the young scholar's passion for science. From a young age, Dr Yang performed well in scientific subjects at school, and her curiosity to explore the frontiers of electronic and communications engineering drove her to immerse herself further into a career in science and research.
Her research journey in the field of health informatics began when she was studying for a doctoral degree in computing science at Imperial College London in the UK. At Imperial, she started developing computational methods for analysing the data generated by medical research. Later as a research associate and then a research fellow at the Data Science Institute at Imperial, she took part in various collaborative international projects that focused on health and medical research. In the U-BIOPRED (Unbiased Biomarkers for the Prediction of Respiratory Disease Outcomes) project, which was supported by the Innovative Medicines Initiative in Europe, Dr Yang and her colleagues used data analysis and machine learning methods to study the data gathered from asthmatic patients, and they provided new insights into severe asthma subtypes.
Dr Yang has also worked as a researcher at Microsoft Research Asia, where she conducted research into the use of AI in cloud computing. Attracted by the research opportunities in academia, she joined HKBU in 2020 and is a member of the University’s System Health Laboratory.
A collaborative research culture
As an early career researcher, Dr Yang is impressed by HKBU's efforts in fostering interdisciplinary research collaborations. She has recently worked with researchers from the School of Communication on a study on vaccine hesitancy in Hong Kong, making use of her data analysis skills to investigate the factors underlying people's reluctance to get vaccinated.
She also appreciates the culture of peer support at the University. "Faculty members often discuss and share ideas on teaching and research, from writing proposals to looking for opportunities to collaborate with colleagues from other faculties and departments. Having this kind of exchange among peers really helps," she says.
With HKBU being selected as the contractor for the service deed of the city's first Chinese Medicine Hospital, Dr Yang hopes that after the hospital's operation commences, she can contribute her expertise in medical AI and data analytics to improve healthcare in Hong Kong.