Predict your future health at the touch of a button

Scientists from Edith Cowan University (ECU) have harnessed the power of artificial intelligence (AI) to revolutionise health predictions, allowing individuals to assess their risk of developing serious health conditions later in life at the mere press of a button. This cutting-edge AI technology can predict the likelihood of individuals developing cardiovascular diseases, falls, fractures, and late-life dementia based on the detection of abdominal aortic calcification (AAC), a condition known to be a major risk factor for such health problems.

AAC occurs when calcium deposits build up within the walls of the abdominal aorta, and it serves as a reliable indicator of future cardiovascular health issues, including heart attacks and strokes. The condition is also linked to an increased risk of falls, fractures, and late-life dementia.

This incredible advancement in AI-driven health predictions marks a significant step forward in proactive healthcare, empowering individuals to take charge of their well-being and make informed decisions to lead healthier and happier lives in their later years. AI holds the potential to reshape the landscape of preventive medicine, benefiting countless lives worldwide.

Previously, detecting AAC required highly trained expert readers to analyze bone density machine scans, a process that could be time-consuming, taking 5-15 minutes per image. However, the collaboration between ECU's School of Science and School of Medical and Health Sciences has led to the development of a sophisticated AI-driven software capable of analyzing an astonishing 60,000 images in a single day.

This tremendous leap in efficiency has the potential to pave the way for widespread use of AAC assessment in research and routine clinical practice. Associate Professor Joshua Lewis, a Heart Foundation Future Leader Fellow and one of the researchers involved in the project, emphasized the significance of this advancement in predicting and preventing health problems later in life.

The international collaboration between ECU, the University of WA, University of Minnesota, Southampton, University of Manitoba, Marcus Institute for Aging Research, and Hebrew SeniorLife Harvard Medical School enabled the study to become the largest of its kind. It was based on the most commonly used bone density machine models and was the first to be tested in a real-world setting using images obtained during routine bone density testing.

The study compared the software's AAC assessments to those made by human experts. Remarkably, the software and expert readers reached the same conclusion about the extent of AAC (low, moderate, or high) in 80 percent of cases, a promising result for the software's first version. Notably, only 3 percent of individuals with high AAC levels were misdiagnosed as having low levels by the software. Identifying these high-risk individuals is crucial as they face a greater likelihood of experiencing fatal and nonfatal cardiovascular events and all-cause mortality.

Although the researchers acknowledge the need for further improvement to match human accuracy levels, they are already working on refining the software with more recent versions. The current AI algorithm opens up the possibility of large-scale screening for cardiovascular disease and other conditions, even before symptoms manifest, enabling individuals at risk to make necessary lifestyle changes early on and promoting better long-term health.

The Heart Foundation's generous funding, facilitated by Professor Lewis' 2019 Future Leadership Fellowship, has been instrumental in supporting this groundbreaking project over a three-year period. The research paper titled 'Machine Learning for Abdominal Aortic Calcification Assessment from Bone Density Machine-Derived Lateral Spine Images' has been published in eBioMedicine.