A silent killer, an abdominal aortic aneurysm is caused by a weakness in the wall of the main artery leaving the heart. Bulges of 5-6 centimetres wide cause ruptures that are abrupt and fatal. Through a precision screening trial in New Zealand, 800 patients aged 60-79 were identified through data analysis of patient records as likely candidates for AAA.
After a simple ultrasound scan for the 632 patients found to be most at risk of having AAA, 36 aneurysms were identified. One man had a 6cm bulge in his aorta and needed immediate surgery. The precision screening trial saved his life.
As healthcare costs continue to rise, healthcare organisations are looking for ways to keep patients out of hospital through prediction and early intervention of health issues before it’s too late. The growing movement towards precision health uses all available data on a patient to generate valuable insights for clinicians such as disease prediction and risk.
The trial was part of a research project under Precision Driven Health, a public-private partnership between Orion Health, Waitemata District Health Board and the University of Auckland that uses data science to develop innovative tools that improve health. The algorithm used is based on their previous epidemiological studies that identified men and women who are most at risk from thousands of patient records. The application of data science and machine learning, specifically precision screening, has the potential to save more lives by identifying a serious health issue before it’s too late.
Developments in predictive analysis like the AAA trial are contributing to a global shift towards precision health, with the goal of improving health outcomes and the effectiveness of interventions.
Wearable device data is a simple yet valuable source for data analysis in health because of its prevalence. The Apple Watch has a built-in app that can measure a person’s heart rate at the time of resting, walking, breathing and exercising on a daily basis. When integrated with other apps that collect nutrition, activity and sleep data, this information paints a clear picture of an individual’s health.
One of the biggest challenges of precision health is: how do we make sense of all this data and use it to make a real impact on improving health outcomes? The reality is, humans – clinicians or the general public – cannot hope to be able to process the masses of data being collected, as well as all knowledge of health. Machine learning will play a crucial role in precision health because models can be trained to recognise, analyse and make predictions at a rate far more advanced than a person could.
Data scientists are training machines to process the massive amount of health data becoming available and make predictions based on the data. If we are able to predict a health issue, it provides the opportunity to prevent it from getting worse and treat it early enough to mitigate the impact on the patient but also the health system.
For the group of patients found to have AAA in the precision screening trial, clinicians were able to intervene early and put patients on the path to a positive recovery rather than waiting until it’s at a life-threatening stage. Precision health leverages the capability to capture and understand health data from entire populations and use it to make predictions on an individual patient.
Precision screening for AAA is one of over 40 research projects supported by Precision Driven Health, applying data science to tackle health challenges.
The Precision Driven Health partnership combines expertise from health, academic and commercial sectors by partnering with clinicians, researchers and health IT organisations. We believe cross-sector collaboration is fundamental in order to bring these innovative tools closer to the people they are trying to help. Our research projects combine clinicians’ needs with the latest and greatest in data science, turning this into sustainable solutions for the healthcare industry.
Dr Kevin Ross is the General Manager of Precision Driven Health and Director of Research at Orion Health.
He will be speaking at HIC in Sydney at the Today’s Best Practice session:
Monday 30 July 2018 - 11:10am-12:40pm
Industry session: The unique patient: Genomics and the arrival of personalised medicine
Precision Driven Health research partnership: Optimising the health of individuals through the use of data