There’s a new, glimmering frontier in healthcare. Artificial Intelligence (AI) and Machine Learning are reshaping the future, with the potential to relieve the burden on health professionals and allow physicians to connect better with patients.
Also known as ‘Deep Medicine’, AI and Machine Learning can help cut down the cost of healthcare. From transforming note taking and scans to saving lives through better diagnoses and treatment, it can free professionals up from time-consuming tasks that interfere with real medicine.
Yet all revolutions come with caveats. AI and Machine Learning also raise a host of meaty issues around governance, data, privacy and legality, plus the overarching question of where ultimate responsibility lies.
The 3rd Annual AI & Machine Learning in Healthcare Conference in Melbourne this November will tackle these issues, as well as showcase its potential and implementation across various clinical settings.
It will look at how AI and Machine Learning are working in practice around the world and what the future holds. Dr Louise Sun will be delivering the Opening International Keynote at the conference, discussing how her team are harnessing data to its full potential to enable the clinical application of AI in Canada.
Dr Sun is a Cardiac Anaesthesiologist, AI Epidemiologist and Director of Big Data and Health Bioinformatics Research at the University of Ottawa Heart Institute (UOHI).
She has been a pioneer in harnessing the power of Big Data, AI and Machine Learning in cardiovascular research. For example, it was the UOHI team that made the seminal discovery in 2018 that death rates from heart failure are higher in women.
Fusing Biomechanics, Fluid Mechanics and Applied Mathematics in Medicine
Dr Sun’s background illustrates the transformative impact of STEM exposure in childhood. Coming from a family of mathematicians and engineers, the young Louise was literate in computer programming from an early age.
Commencing her undergraduate studies in biomedical engineering, Dr Sun was encouraged to apply to medical school during an internship with the National Research Council of Canada, where she developed a software package to fine tune the 3D reconstruction of brain tumour images to improve the precision of radiotherapy.
“As a medical student, I became interested in human factor engineering in the operating room as well as cardiovascular physiology, which is a real-life application of biomechanics, fluid mechanics and applied mathematics. These interests led me to pursue subspecialty training in Cardiac Anaesthesiology.”
Dr Sun is now at the vanguard of using AI and Machine Learning techniques to model disease progression and predict outcomes after cardiac procedures.
“We start with clinically and patient-relevant questions,” says Dr Sun. “These questions often come up during hallway conversations with physician colleagues and we distil them down into solvable subcomponents afterwards.”
Examples of the team’s work include the prediction of death and complications on the surgical waiting list, prediction of disability-free survival as a patient-centred outcome after open heart surgery and minimally invasive cardiac procedures, creation of a smart alert system for critical deterioration in the ICU, and prediction of the onset and worsening of chronic diseases in the ambulatory care setting.
Notably, SMART medicine has also helped Dr Sun’s team navigate the challenges of the pandemic, which has caused worldwide delays to surgery.
“One practical example is our rapid development and implementation of a set of digital decision tools to help prioritise high-risk patients for cardiac surgery and to improve ICU and hospital throughout.”
“We’ve used some of these tools to triage patients since the onset of the pandemic. While other centres reduced their annual procedure volumes, we managed to perform close to our usual cardiac surgery volumes while still keeping enough ICU beds open for COVID-19.”
‘Buy In’ is the Key to Unlocking the Potential of Deep Medicine
Dr Sun acknowledges that it can be difficult to engage all hospital stakeholders with AI and Machine Learning. However, she underscores that it is fundamental to success.
“Having clinician data scientists as leaders and champions, and having open-minded administrators onboard are essential to our success.”
“Our digital cardiac surgery triage tools would not have been a success without buy-in and collective effort of our surgery and anaesthesia colleagues, our triage coordinator, clinical operations team, IT, and support of our senior administration.”
The future of AI and Machine Learning is an exciting one, although one that Dr Sun says needs to be closely monitored.
“I see a lot of potential in routinely collected data. This could be EHR or healthcare administrative datasets. Advances in data governance and security are making it easier to access this data for training and validating clinically relevant algorithms, which in turn could be embedded in these systems for continuous disease surveillance, as well as evidence-based, patient-centred decision-making without interrupting the clinical workflow.
“A caveat is that for a model to be useful, it needs to be developed and validated in the population/healthcare setting where it’s intended to be used. Models also need to be constantly updated to adapt to temporal changes in patient and provider demographics, amongst other variables.”
Don’t miss Dr Louise Sun’s valuable insights at the 3rd Annual AI and Machine Learning in Healthcare Conference, 16 -17 November 2021 at the Rendezvous Hotel in Melbourne.
This event is unmissable for key healthcare stakeholders from across the spectrum. It will focus on a variety of clinical settings and be attended by heads of innovation, clinicians, digital health and transformation managers, health informaticians and healthcare IT professionals. View the full agenda and book here.